Google Cloud VMware Engine (GCVE)

Google Cloud VMware Engine (GCVE)

In June 2020 VMware and Google made the announcement that Google Cloud VMware Engine (GCVE) is generally available. Almost exactly one year ago, the market received the information that VMware’s Cloud Foundation (vSphere, vSAN and NSX) stack will come to Google Cloud.

With this milestone VMware is now present on top of all the so-called “big three” hyperscalers.

GCVE has the same goals like the other similar offerings like VMware Cloud on AWS or Azure VMware Solution and belongs to the VMware multi-cloud strategy – to seamlessly migrate and run applications in the public cloud. In this case in Google Cloud! Run your applications in the public cloud exactly the same way as you already do now withh your on-premises VMware environment. With the very important addition, that you have high speed access to Google Cloud services like Cloud SQL, Cloud Storage, big data or AI/ML services.

To be able to run VMware workloads on top of the Google Cloud global infrastructure, Google acquired CloudSimple (with which they partnered with already) last November 2019.

At the moment of writing, the VMware hybrid cloud experience on Google Cloud is sold, operated and supported by Google and their partners.

Many customers are already looking at this very interesting offer, which is going to be available in more regions until the end of 2020. But there are also already a few customers using the joint offering. Google just published a customer reference story about the “Deutsche Börse Group”, a large and international financial organization, which extended their on-premises environment to Google Cloud with Google Cloud VMware Engine. One of the reasons why Deutsche Börse went for this vSphere-based cloud approach, was, to keep migrations to the cloud easy. I expect we can hear more about this success story at VMworld 2020.

Cloud Migration and Workload Mobility

A lot of customers underestimate the amount of work, time and costs involved in refactoring or re-platforming applications and the overall challenges when it comes to migrations from on-prem to the cloud. To build this secure hybrid cloud extension with GCVE, you’ll need VMware HCX, which is included in the GCVE offering.

There are different options available to connect both worlds:

GCVE Connectivity Options

  • VPN Gateway for point-to-point connections, used for the secure admin access to vCenter. Useful for the initial setup of the GCVE environment.
  • Cloud VPN for site-to-site connections, a secure layer 3 connection over the internet. This is one of the lower cost options for use cases, that don’t require high bandwidth.
  • Dedicated Cloud Interconnect with a direct traffic flow to Google with 10Gbps or 100Gbps circuits with 50Mbps to 50 Gbps connection capacities. This direct connection is required for HCX and the preferable connectivity option for customers requiring high speed and low latency.
  • Partner (Cloud) Interconnect is another option of a Cloud Interconnect, where your traffic flows through one of the supported service providers (e.g. Colt, Equinix, BT, e-shelter, Verizon, InterCloud, Interxion, Megaport)

Note: One unique feature of GCVE is the ability to route between different GCVE environments in the same region, without the need for additional configuration. 

Use Cases

These use cases, if you made yourself already familiar with a hybrid cloud approach, shouldn’t be new to you.

Data Center extension or retirement. You can scale the data center capacity in the cloud on-demand, if you for example don’t want to invest anymore in your on-premises environment. In case you just refreshed your current hardware, another use case would be the extension of your on-premises vSphere cloud to Google Cloud.

Disaster Recovery and data protection. Here we’ll find different scenarios like recovery (replication) or backup/archive (data protection) use cases. You can also still use your existing 3rd party tools from Zerto or Veeam to replace or complement existing DR locations and leverage the Cloud Storage service. You can also use your GCVE private clouds as a disaster recovery (DR) site for your on-premises workloads. This DR solution would be based on VMware Site Recovery Manager (SRM) which can be also used together with HCX.

Cloud migrations or consolidation. If you want to start with a lift & shift approach to migrate specific applications to the cloud, then GCVE is definitely right for you. Maybe you want to refresh your current infrastructure and need to relocate or migrate your workloads in an easy and secure way? Another perfect scenario would be the consolidation of different vSphere-based clouds.

Application modernization. Re-architecting or refactoring applications is not that easy. Most customers start with a partial approach to modernize their applications and leverage cloud-native services (e.g. databases, AI/ML engines).

Interesting: Did you know that Google’s on-prem GKE (Google Anthos) is running on vSphere?

VMware Horizon on VMware Engine

The advantages of a public cloud like Google Cloud are the “endless” capacity, agility and high-bandwidth connections. These items are very important for a virtual desktop infrastructure (VDI) and specially during disaster scenarios, when onboardings have to happen fast or if you look for on-demand growth.

Another regular example could be a merger & acquisition use case, where we the main infrastructure doesn’t have the necessary physical resources to onboard to new company and their employees.

Because something like this has always happen as easy and fast as possible. Running virtual desktops in Google Cloud VMware Engine can help in such situations. Together with VMware Horizon, organizations could install a VDI environment in GCVE and connect it to their Horizon on-premises infrastructure using the Cloud Pod Architecture (CPA). 

Note: When migrating applications to the cloud (GCVE), it is a best practice to keep the virtual desktop close to the application, which is a general use case we see when talking about application locality.

Horizon Global Pod GCVE

With the release of Horizon 2006 (aka Horizon 8) it is also possible to choose “Google Cloud” as deployment option during the connection server installation.

C:\Users\mrebmann\OneDrive - VMware, Inc\cloud13\2020 - Google Cloud VMware Engine\Horizon on GCVE.png

In case you need a load balancer (for your Horizon components and in general) for your on-premises environment and the public cloud, have a look at NSX Advanced Load Balancer.

GCVE Node Specs

When planning your GCVE resource needs, be aware of the following specifications and limits:

CPU: Intel Xeon Gold 6240 (Cascade Lake) 2.6 GHz (x2), 36 Cores, 72 Hyper-Threads

Memory: 768 GB

Storage (vSAN): 2 × 1.6 TB (3.2 TB) NVMe (Cache), 6 × 3.2 TB (19.2 TB) NVMe (Data)

Number of nodes required to create a private cloud: 3 (up to 64 hosts per private cloud)

Number of nodes allowed in a cluster on a private cloud: 16

3rd party tools compatibility: Yes, you can use existing tools (elevated privileges allow you to install 3rd party software)

Interesting facts: It only takes about a half hour to spin up your private cloud with three nodes! The addition of a new node takes approximately 15 minutes.

GCVE Elevated Privileges

Software License and Versions

Please find the current software versions and licenses below used for the GCVE offering (purchased with a 1- or 3- year commitment). The listed software versions are fixed and all updates are managed by Google. Google is responsible for the lifecycle management of the VMware software, which includes ESXi, vCenter and NSX.

Component Version License
vCenter 6.7 U3 vCenter Standard
ESXi 6.7 U3 Enterprise Plus
vSAN 6.7 U3 Enterprise
NSX Data Center (NSX-T) 2.5.1 Advanced
HCX 3.5.3 Advanced

Shared Responsibilities

Google Cloud VMware Engine is coming with all components you need to securely run VMware natively in a dedicated private cloud. Google takes care of the infrastructure (service) and their native service integrations. As a customer you only need to take care of your virtual machines or containers with your applications and data. Besides that, you also need to make sure that your configurations, policies, network portgroups, authentication and capacity management are properly configured.

GCVE Shared Responsibilities

If you want to know and learn more about Google Cloud VMware Engine, have a look at the following resources: 

Multi-Cloud Load Balancing and Autoscaling with NSX Advanced Load Balancer (formerly Avi Networks)

Multi-Cloud Load Balancing and Autoscaling with NSX Advanced Load Balancer (formerly Avi Networks)

Do you want to build your private cloud like a hyperscaler is doing it? You know that VMware Cloud Foundation is becoming the new vSphere, but still wonder how you can implement software-defined load balancing (LB) or application services and features like autoscaling or predictive scaling? Then this article about multi-cloud load balancing and autoscaling with NSX Advanced Load Balancer aka Avi Networks is for you!

My Experience with a Legacy ADC

A few years ago, I was working on the customer side for an insurance company in Switzerland as a Citrix System Engineer. My daily tasks included the maintenance and operation of the Citrix environment, which included physical and virtual Citrix NetScaler Application Delivery Controller (ADC) appliances. The networking team owned a few hardware-based appliances (NetScaler SDX) with integrated virtualization capability (XenServer as hypervisor) to host multiple virtual NetScaler (VPX) instances.

The networking team had their dedicated NetScaler VPX instances (for LDAP and HTTP load balancing mostly) and deployed my appliances after I filed a change request. Today, you would call this multi-tenancy. For a Citrix architecture is was best practices to have one high availability (HA) pair for the internal and one HA pair for the external (DMZ) network access. A HA pair was running in a active/passive mode and I had to maintain the same setup for the test environment as well.

Since my virtual VPX appliances were hosted on the physical SDX appliance, I always relied on the network engineers, if I needed more resources (CPU, RAM, SSL, throughput) chips allocated to my virtual instances. Before I could upgrade to a specific firmware version, I also had to wait until they upgraded the physical NetScaler appliances and approved my change request. This meant, we had to plan changes and maintenance windows together and had to cross fingers, that their upgrade went well, that we could upgrade all our appliances after.

NetScaler SDX

It was also possible to download a VPX appliance, which could run on top of VMware vSphere. To be more independent, I decided to install four new VPX appliances (for the production and test environment) on vSphere and migrated the configuration from the appliances running on the physical SDX appliance.

Another experience I had with load balancers was when I started to work for Citrix as a consultant in Central Europe and had to perform a migration of physical NetScaler MPX appliances, which had no integrated virtualization capability. I believe I had two sites with each two of these powerful MPX appliances for tens of thousands of users. Beside the regular load balancing configuration for some of the Citrix components, I also had to configure Global Server Load Balancing (GSLB) in active/passive mode for the two sites.

NetScaler GSLB Active Passive

There were so many more features available (e.g. Web Application Firewall, Content Switching, Caching, Intrusion Detection), but I never used anything else than the NetScaler Universal Gateway for the remote access to the virtual desktop infrastructure (VDI), load balancing, HTTP to HTTPS redirections and GSLB. In all scenarios I had a HA pair where one instance was idling and doing nothing. And the active unit was in average not utilized more than 15-20%. It was common to install/buy too large or powerful instances/licenses, because you wanted to be on the safe side and have enough capacity to terminate all your SSL sessions and so on.

It (load balancing) was about distributing network traffic across multiple servers by spreading the requests and work evenly, and do add some intelligence (health monitoring) in case an application server or a service would fail or be unavailable for any reason. If one more application server was needed, I ordered a new Windows Server, installed and configured the Citrix components and added the necessary load balancing configuration on the NetScaler. These were all manual tasks. The same work has been done by the network engineers when the application team requested a new application server, which then had to be added to the load balancing configuration on their NetScaler appliances.

This was my personal experience from 2017. Since then applications became more complex and distributed. The analysts and market are focusing on containerized and portable apps running and more and more in multiple clouds. The prediction is also that the future is multi-cloud.

Multiple Clouds vs. Multi-Cloud

There are different definitions and understandings out there what multi-cloud means. In my understanding, using a private cloud, AWS, Microsoft Office 365 and Azure are a typical setup with multiple clouds. There are simple scenarios where you migrate workloads from the private to the public cloud (e.g. Azure) or having applications with services lying on the private on the public cloud. The latter would be an example of a hybrid cloud architecture.

The reasons for which services and resources are needed or distributed on multiple clouds (on-prem, Azure, AWS, GCP etc.) are various:

  • Avoid dependence on only one cloud provider
  • Consume different specific services that are not provided elsewhere
  • Optimize costs for different workloads and services
  • React to price changes by the providers

That is why we are seeing also the trend to break up big legacy applications (monoliths) in smaller pieces (segments), which is a best practice and design principle today. The goal is to move to a loosely coupled and more service-oriented architecture. This provides greater agility, more flexibility and easier scalability, because of less inter-dependencies.

And, if we take the second example from the list before, a segmented application is much easier to run in different clouds (portability). Running one application over multiple clouds is in my understanding the right definition of multi-cloud.

Multiple Clouds versus Multi-Cloud

Let’s assume that most probably all the four reasons above apply to larger enterprises. If we take another angle, we can define some business and technical requirements for multi-cloud:

  • Application or services need to be cloud vendor-agnostic
  • Provide or abstract control and management interface of multiple clouds
  • Support application portability/relocation between clouds
  • Combine IaaS and services from different clouds
  • Possibility to deploy components of applications in multiple clouds
  • (Cloud) Broker service needed
  • Policy and governance over multiple clouds
  • Network connectivity for migration scenarios with partially modernized applications
  • Automated procedures for deployments
  • Application monitoring over different clouds
  • Costs management
  • Lifecycle management of deployed applications in multiple clouds
  • Self-adaption and auto scaling features
  • Large team with various expertises needed

How can you deliver and manage the different applications services like load balancing, web application firewalling, analytics, automation and security over multiple clouds?

Another important question would be, how you want to manage the deployment on the various clouds. But cloud management or a cloud management platform is something for another article. 🙂

The requirements for the developers, operations and the business are very complex and it’s a long list (see above).

It is important, that you understand, based on the requirements for multi-cloud, that it is mandatory to implement a modern solution for your modernized application architectures. Enterprises have become more application-centric and everyone is talking about continious integration, continuous delivery and DevOps practices to automate operation and deployment tasks. A modern solution implicits a software-defined approach. Otherwise you won’t be able to be agile, adapt to changes and meet future requirements.

My past experience with Citrix’ NetScaler is a typical example that “virtualized” and “software-defined” are not the same thing. And this is very important if we want to have a future-ready solution. If we look at VMware’s software-defined data center (SDDC), beside the virtual compute, also includes software-defined storage and networking. Part of the software-defined networking portfolio is “NSX Advanced Load Balancer“, the software-defined application services platform, which was also known as “Avi Networks” before VMware bought them in June 2019.

Unlike a virtualized load-balancing appliance, a software-defined
application services platform separates the data and control planes
to deliver application services beyond load balancing, real-time
application analytics, security and monitoring, predictive autoscaling,
and end-to-end automation for Transport (Layer 4) to
Application (Layer 7) layer services. The platform supports multicloud
environments and provides software-defined application
services with infrastructure-agnostic deployments on bare metal
servers, virtual machines (VMs), and containers, in on-premises
data centers and private/public clouds.

Autoscaling became famous with AWS as it monitors your applications and automatically adjusts capacity to maintain availability and performance at the lowest possible costs. It automatically adds or removes application servers (e.g. EC2 instances), load balancers, applies the right network configuration and so on.

Can you achieve the same for your on-premises infrastructure with VMware? Yes.

Is there even a solution which can serve both worlds – on-prem and cloud? Yes.

And what about predictive scaling with real-time insights? Yes.

NSX Advanced Load Balancer (NSX ALB)

Why did VMware buy Avi? Because it follows the same architecture principles like NSX: A distributed platform with a separate control and data plane built on software-defined principles for any cloud.

Avi High Level Architecture

Traditional ADCs or load balancers are mostly configured in active/standby pairs, no matter if physical or virtual. Typically you would see around 15% utilization on the active node where the secondary standby node is just idling and doing nothing. Each pair is its own island of static capacity which shares the management, control and data plane.

You have to decide where to place the virtual IP (VIP) and how much you want to overprovision the physical or virtual appliances, because there is no capacity pooling available. This leads to operational complexity, especially when you have hundreds of such HA pairs running in different clouds. Therefore, legacy and virtualized ADCs are not the ideal choice for a multi-cloud architecture. Let’s check NSX ALB’s architecture:

Control Plane – This is the brain (single point of management) of the whole platform that can be spun up in your on-prem environment or in the cloud (also available as a managed SaaS offering), typically as a three-node cluster. Within this cluster, all configuration is done, this also where the policies reside and the decisions are made. It is the controller’s duty to place virtual services on SEs to load balance new applications or increase the capacity of running applications.

The control plane comprises the three pillars that deliver the key capabilities of the Avi platform:

  • Multi-Cloud – Consistent experience for any cloud, no lock-in
  • Intelligence – The machine learning based analytics engine enables application performance monitoring, troubleshooting, and operational insights (gathered by the SEs)
  • Automation – Elastic and predictive auto scaling & self-service without over-provisioning through a complete set of REST APIs

Data Plane –  The Service Engines (SEs) handle all data plane operations by receiving and executing instructions from the controller. The SEs perform load balancing and all client- and server-facing network interactions. It collects real-time application telemetry from application traffic flows. 

As already mentioned, NSX ALB can be deployed in multiple cloud environments like VMware vCenter, Amazon Web Services, Microsoft Azure, Google Cloud Platform, Oracle Cloud, IBM Cloud, VMC on AWS, Nutanix, OpenStack or bare-metal.

Use Cases

Most customers deploy Avi because of:

  • Load Balancer refresh
  • Multi-Cloud initiatives
  • Security including WAF, DDoS attack mitigation, achieve compliance (GDPR, PCI, HIPAA)
  • Container ingress (integrates via REST APIs with K8s ecosystems like GKE, PKS (TKGI), OpenShift, EKS, AKS, TKG)

Advanced Kubernetes Ingress Controller Avi Networks

  • Virtual Desktop Infrastructure (Citrix, VMware Horizon)

Consistent Application Services Platform (Features)

Avi/NSX ALB is an enterpise-grade solution. So, everything you would expect from a traditional ADC (e.g. F5), layer 4 to layer 7 services, SSL, DDoS, WAF etc. is built-in without the need for a special license edition. There is also no NSX license requirement even the product name would suggest it. It can be deployed as a standalone load balancer or as an integrated solution with other VMware products (e.g. VCF, vRA/vRO, Horizon, Tanzu etc.).

Avi Networks Features

Below is a list with the core features:

  • Enterprise-class load balancing – SSL termination, default gateway, GSLB, DNS, and other L4-L7 services
  • Multi-cloud load balancing – Intelligent traffic routing across multiple sites and across private or public clouds
  • Application performance monitoring – Monitor performance and record and replay network events like a Network DVR
  • Predictive autoscaling Application and load balancer scaling based on real-time traffic patterns
  • Self-service – For app developers with REST APIs to build services into applications
  • Cloud connectors – VMware Cloud on AWS, SDN/NFV controllers, OpenStack, AWS, GCP, Azure, Linux Server Cloud, OpenShift/Kubernetes
  • Distributed application security fabric – Granular app insights from distributed service proxies to secure web apps in real time
  • SSO / Client Authentication – SAML 2.0 authentication for back-end HTTP applications
  • Automation and programmability – REST API based solution for accelerated application delivery; extending automation from networking to developers
  • Application Analytics – Real-time telemetry from a distributed load balancing fabric that delivers millions of data points in real time

Load Balancing for VMware Horizon

NSX ALB can be configured for load balancing in VMware Horizon deployments, where you place SEs in front of Unified Access Gateways (UAG) or Connection Servers (CS) as required.

Avi Horizon High Level Architecture

For a multi-site architecture you can also configure GSLB if needed. With GSLB, access to resources is controlled with DNS queries and health checking.

Note: If you are using the Horizon Universal Broker, the cloud-based brokering service, there is no need for GSLB, because the Universal Broker can orchestrate connections from a higher level based on different policies.

Automation

With NSX Advanced Load Balancer there are two parts when we talk about automation. One part is about infrastructure automation, where the controller talks to the ecosystem like a vCenter, AWS or Azure to orchestrate the Service Engine. So, when you configure a new VIP, the controller would talk to vCenter to spin up a VM, put it in the right portgroup, connect the front and the back-end, download the policy and service engine, and starts receiveing traffic.

The second piece of automation focuses more on the operational automation which is through the REST API (the UI and CLI don’t offer all the configuration, 100% can be done via REST API). But, on top of that you can also run Ansible playbooks, Terraform templates, Go and Python SDKs, have integrations with Splunk or other tools like vRealize Automation. This is the built-in automation in the product.

Avi Networks Automation

VMworld 2020 Sessions

This year VMworld is going to be for free and virtual. Take this chance and register yourself and learn more about Avi aka NSX ALB:

  1. Making Your Private Cloud Network Run Like a Public Cloud – Part 2 [VCNC2918]
  2. Modern Apps and Containers: Networking and Security [VCNC2920]
  3. Prepare for the New Normal of Work from Anywhere [VCNC2919]

Expectations and Current Approaches

There is the general understanding and need for hybrid or multi-cloud architectures. Different people will tell different stories and give different advices. The result are different architectures and different approaches. Some people will tell you, that you can use a cloud serially, so moving from one cloud to another. Or, simultaneously, when using different services from different clouds.

My last article focused on hybrid cloud, the architecture with some services lying on the private infrastructure, while other services are hosted on a public cloud. A public cloud providers tells you, that you can buy all services from them and tries to give you a better discount than the competition (to avoid multiple clouds). Enterprises see the need for multiple (public) clouds to avoid a vendor lock-in instead of going all-in with just one of them.

VMware is about multi-cloud and workload mobility, with the vision, that their VCF stack is running everywhere in the future. Now, some people would now say that this is also a vendor lock-in. Depending on your strategy and technology choices and preferences (e.g. databases, AI/ML services, virtual desktops), you have to decide somewhen which (cloud) vendor, approach and operation model is the right one for you.

It may not true for every large environment, but if you go for multiple clouds, multiple technologies, management and security consoles, architecture and so on, you’ll spend a lot of time and money on engineering and keeping your environment “integrated” and functional.

VMware offers you choice. The choice to run your workloads today and tomorrow wherever you want.

If you have the same vision and strategy like VMware, then you are looking for solutions which run in or on top of every cloud. Because of that it’s very important to understand the different between multiple clouds and multi-cloud.

In this case, NSX ALB brings you multi-cloud load balancing and auto scaling features for any cloud and for multi-cloud enabled applications and services.

Don’t forget: Some people are also saying,  that multi-cloud is not needed and doesn’t exist in reality. Nobody is saying multi-cloud is a piece of cake, but VMware can definitely help you to abstract this complexity. And part of this abstraction can be handled with vRealize Automation for example, which can act as a cloud broker to deploy your application and services.

 

VMware Multi-Cloud and Hyperscale Computing

VMware Multi-Cloud and Hyperscale Computing

In my previous article Cross-Cloud Mobility with VMware HCX I already very quickly talked about VMware’s hybrid and multi-cloud vision and strategy. I mentioned, that VMware is coming from the on-premises world if you compare them with AWS, Azure or Google, but have the same “consistent infrastructure with consistent operations” messaging. And that the difference would be, that VMware is not only hardware-agnostic, but even cloud-agnostic. To abstract the technology format and infrastructure in the public cloud, their idea is to run VMware Cloud Foundation (VCF) everywhere (e.g. Azure VMware Solution), on-premises on top of any hardware and in the cloud on any global infrastructure from any hyperscaler like AWS, Azure, Google, Oracle, IBM, Alibaba. Or you can run your workloads in a VMware cloud provider’s cloud based on VCF. That’s the VMware multi-cloud.

The goal of this article is not compare any features from different vendors and products, but to give you a better idea why multi-cloud is becoming a strategic priority for most enterprises and why VMware could be right partner for your journey to the cloud.

To get started, let’s get an understanding what the three big hyperscalers are doing is when it comes to a hybrid or multi-cloud.

Microsoft

To bring Azure services to your data center and to benefit from a hybrid cloud approach, you would probably go for Azure Stack to run virtualized applications on-premises. Their goal is to build consistent experiences in the cloud and at the edge, even for scenarios where you have no internet connection. This would be by VMware’s definition a typical hybrid cloud architecture.

Multi-cloud refers to the use of multiple public cloud service providers in a multi-cloud architecture, whereas hybrid cloud describes the use of public cloud in conjunction with private cloud. In a hybrid cloud environment, specific applications leverage both the private and public clouds to operate. In a multi-cloud environment, two or more public cloud vendors provide a variety of cloud-based services to a business.

With the announcement of Azure Arc at MS Ignite 2019, Microsoft introduced a new product, which “simplifies complex and distributed environments across on-premises, edge and multi-cloud“. Beside the fact that you can run Azure data services anywhere, it gives you the possibility to govern and secure your Windows servers, Linux servers and Kubernetes (K8s) clusters across different clouds. Arc can also deploy and manage K8s applications consistently (from source control).

Azure Arc InfographicYou could summarize it like this, that Microsoft is bringing Azure infrastructure and services to any infrastructure. It’s not necessary to understand the technical details of Azure Stack and Azure Arc. More important is the messaging and the strategy. It’s about managing and securing Windows/Linux servers, virtual machines and K8s clusters everywhere and this with their Azure Resource Manager (ARM). Arc ensures that the right configurations and policies are in place to fulfill governance requirements across clouds. Run your workloads where you need it and where it makes sense, even it isn’t Azure.

Google Anthos

Google open-sourced their own implementation of containers to the Linux kernel in about 2006 or 2007. It was called cgroups, which stands for control groups. Docker appeared in 2013 and provided some nice tooling for containers. Over the next years, Microservices were used more often to divide monoliths into different pieces and services. Because of the growing numbers of containers, Google saw the need to make this technology easy to manage and orchestrate for everyone. This was six years ago when they released Kubernetes.

By the way, two of the three Kubernetes founders, namely Joe Beda and Craig McLuckie, are working for VMware since their company Heptio has been acquired by VMware in November 2018.

Today, Kubernetes is the standard way to run containers at scale.

We know by now that the future is hybrid or even multi-cloud, and not public cloud only. Also Google realized that years ago. Besides that, a lot of enterprises made the experience that moving to the cloud and re-engineering the whole application at the same time mostly fail. This means, that moving applications from your on-premises data center, refactoring the application at the same time and run it in the public cloud, is not that easy.

Why isn’t it easy? Because you are re-engineering the whole application, have to take care of other application and network dependencies, think about security, governance and have to train your staff to cope with all the new management consoles and processes.

Google’s answer and approach here is to modernize applications on-premises and then move them to the cloud after the modernization happened. They say that you need a platform, that runs in the cloud and in your data center. A platform, that runs consistently across different environments – same technology, same tools and policies everywhere.

This platform is called Google Anthos. Anthos is 100% software-defined and (hardware) vendor-agnostic. To deliver their desired developer experience on-prem as well, they rely on VMware. This is GKE running on-prem on top of vSphere:

Anthos vSphere on-prem

Amazon Web Services

The last solution I would like to mention is AWS Outposts, which is a fully managed service that extends their AWS infrastructure, services and tools to any data center for a “truly consistent hybrid experience”. What are the AWS services running on Outposts?

  • Containers (EKS)
  • Compute (EC2)
  • Storage (EBS)
  • Databases (Amazon RDS)
  • Data Analytics (Amazon EMR)
  • Different tools and APIs

AWS Outposts are delivered as an industry-standard 42U rack. The Outpost rack is 80 inches (203.2cm) tall, 24 inches (60.96cm) wide, and 48 inches (121.92cm) deep. Inside we have hosts, switches, a network patch panel, a power shelf, and blank panels. It has redundant active components including network switches and hot spare hosts.

If you visit the Outposts website, you’ll find the following information:

Coming soon in 2020, a VMware variant of AWS Outposts will be available. VMware Cloud on AWS Outposts delivers a fully managed VMware Software-Defined Data Center (SDDC) running on AWS Outposts infrastructure on premises.

VMC on AWS Outposts is for customers, who want to use the same VMware software conventions and control plane as they have been using for years. It can be seen as an extension from the regular VMC on AWS offering which is now made available on-premises (on top of the AWS Outposts infrastructure) for a hybrid approach.

VMC on AWS Outposts

What do all these options have in common? It is always about consistent infrastructure with consistent operations. To have one platform in the cloud and on-premises in your data center or at the edge. Most of today’s hybrid cloud strategies rely on the facts, that migrations to the cloud are not easy, fail a lot and so it’s clear why we still have 90% of all workloads running on-premises. We are going to have many million containers more in the future, which need to be orchestrated with Kubernetes, but virtual machines are not just disappearing or being replaced tomorrow.

My conclusion here is, that every hyperscaler is seeing cloud-native in our (near) future and wants to provide their services in the cloud and on-prem. That customer can build their new applications with a service-oriented architecture or partially modernize existing monoliths (big legacy applications) on the same technology stack.

Consistent Infrastructure & Consistent Operations

All hyperscalers mention as well, that you have to take care of different management and security consoles, skills set and tools in general. Except Microsoft with Azure Arc, nobody else is having a “real” multi-cloud solution or platform. I want to highlight, that even Azure Arc is only here for some servers, Kubernetes clusters and takes care of governance.

Let’s assume you have a hybrid cloud setup in place. Your current project requirements tell you to develop new applications in the Google Cloud using GKE. That’s fine. Your current on-premises data centers run with VMware vSphere for virtualization. Tomorrow, you have to think about edge computing for specific use cases where AI and ML-based workloads are involved. Then you decide to go for Azure and create a hybrid architecture with Azure Stack and Arc. Now you are using two different public cloud providers, one with their specific hybrid cloud offering and also VMware vSphere on-premises.

What are you going to do now? How do you manage and secure all these different clouds and technologies? Or do you think about migrating all the application workloads from on-prem to GCP and Azure? Or do you start with Anthos now for other use cases and applications? Maybe you decide later to move away from VMware and evacuate the VMware-based private cloud to any hyperscaler? Is it even possible to do that? If yes, how long would this technology change and migration take?

Let’s assume for this exercise, that this would be a feasable option with an acceptable timeframe. How are you going to manage the different servers, applications, dependencies and secure everything at the same time? How can you manage and provision infrastructure in an easy and efficient way? What about cost control? What happens if you don’t see Azure as strategic anymore and want to move to AWS tomorrow? Then you figure out, that cloud is more expensive than you thought and experience yourself why only 10% of all workloads are running in the public cloud today.

VMware Multi Cloud Silos

I think people can pretty easy handle an infrastructure which runs VMware on-premises and have maximum one public cloud only – a hybrid cloud architecture. If we are talking about a greenfield scenario where you could start from scratch and choose AWS including AWS Outposts, because you think it’s best for you and matches all the requirements, go for it. You know what is right for you.

But I believe, and this is also what I see with larger customers, the current reality is hybrid and the future is multi-cloud.

VMware Multi-Cloud Strategy

And a multi-cloud environment is a totally different game to manage. What is the VMware multi-cloud strategy exactly and why is it different?

Unified Cloud with VMware

VMware’s approach is always to abstract complexity. This doesn’t mean that everything is getting less complex, but you will get the right platform and tooling to deal with this complexity.

A decade ago, abstracting meant providing a hypervisor (vSphere) for any hardware (being vendor-agnostic). After that we had software-defined storage (vSAN) followed software-defined networking (NSX). Beside these three major software pieces, we also have the vRealize suite, which is mainly known for products like vRealize Automation and vRealize Operations. The technology stack consisting of vSphere, vSAN, NSX, vRealize and some management components form the software-defined data center and is called VMware Cloud FoundationA technology stack that allows you to experience the ease of public cloud in your data center. Again, if wanted and required, you can run this stack on top of any hyperscaler like AWS, Azure, Google Cloud, Alibaba Cloud, Oracle Cloud or IBM.

VMware Cloud Foundation

It’s a platform which can deliver services as you would expect in the public cloud. The vRealize suite can help you to automatically provision virtual machines and containers including the right network and storage (any vSphere-based cloud or cloud-native on AWS, GCP, Azure or Alibaba). Build your own templates or blueprints (Infrastructure as Code) to deliver services IaaS, DBaaS, CaaS, DaaS, FaaS, PaaS, SaaS and DRaaS, which can be ordered and consumed by your users or your IT. Put a price tag behind any service or workload you deploy, and include your public cloud spending as well (e.g. with CloudHealth) in this calculation.

You want to deliver vGPU enabled virtual machines or containers? Also possible with vSphere. Modern AI/ML based applications need compute acceleration to handle large and complex computation. vSphere Bitfusion allows you to access GPUs in a virtualized environment over the network (ethernet). Bitfusion works across any cloud and environment and can be accessed from any workload from any network. This topic gets very interesting if we talk about edge computing for example.

vSphere Bitfusion

Modern applications obviously demand a modern infrastructure. An infrastructure with a hybrid or multi-cloud architecture. With that you are facing the challenge of maintaining control and visibility over a growing number of environments. In such a modern environment, how do you automate configuration and management? What about networking and security policies applied at a cluster level? How you handle identity and access management (IAM)? Any clue about backup and restore? And what would be your approach for cost management in a multi-cloud world?

Modern Applications Challenges

To improve the IT ops and developer experience, VMware announced the Tanzu portfolio including something they call the Tanzu Kubernetes Grid (TKG). The promise of TKG is to provide developers a consistent and on-demand access to infrastructure across clouds and is considered to be the enterprise-ready Kubernetes runtime.

Since vSphere 7, TKG has been embedded into the control plane vSphere 7 with Kubernetes as a service. Finally, as Kubernetes is natively integrated into the hypervisor, we have a converged platform for VMs and containers. IT ops now can see and manage Kubernetes objects (e.g. pods) from the vSphere client and developers use the Kubernetes APIs to access the SDDC infrastructure.

There are different ways to consume TKG beside “vSphere 7 with Kubernetes“. TKG is a consistent and upstream compatible Kubernetes runtime with preintegrated and validated components, that also runs in any public cloud or edge environments.

If you have to run Kubernetes clusters natively on Azure, AWS, Google and on vSphere on-premises, how would you manage IAM, lifecycle, policies, visibility, compliance and security? How would you manage any new or existing clusters?

Tanzu Mission Control

Here, VMware’s solution would be Tanzu Mission Control (TMC). A centralized management platform (operated by VMware as SaaS) for all your clusters in any cloud. TMC allows you to provision TKG workload clusters to your environment of choice and manage the lifecycle of each cluster via TMC. To date, the supported deployments are in vSphere and AWS EC2 accounts. The deployment on Azure is coming very soon.

Existing Kubernetes clusters from any vendor such as EKS, AKS, GKE or OpenShift can be attached to TMC. As long as you are maintaining CNCF conformant clusters, you can attach them to TMC so that you can manage all of them centrally.

The Tanzu portfolio is much bigger and includes more than TKG and TMC, which only address the “where and how to run Kubernetes” and “how to deploy and manage Kubernetes”. Tanzu has other solutions like an application catalog, build service, application service (previously Pivotal Cloud Foundry) and observability (monitoring and metrics) for example.

VMware Tanzu Products

And this Tanzu products can be complemented with cloud-scale networking solutions like an application delivery controller (ADC) or software-defined WAN (SD-WAN). To deliver the “public cloud experience” to developers for any infrastructure, we need to provide agility. From an infrastructure perspective we’ll find VMware Cloud Foundation and from application or developer perspective we learned that Tanzu covers that.

For a distributed application architecture, you also need a software-defined ADC architecture that is fully distributed, auto scalable and provides real-time analytics and security for VMs or containers. VMware’s NSX Advanced Load Balancer (formerly known as Avi Networks) runs on AWS, GCP, Azure, OpenStack and VMware and has a rich feature set:

AVI Networks Features

Hypervisor versus Public Cloud

What I am trying to say here, is, that cloud-native at scale requires much more than containers only. While hypervisors are obviously not disappearing and getting replaced by containers from the public cloud very soon, they will co-exist and therefore it is very important to implement solutions which can be used everywhere. If you can ignore the cost factor for a moment, probably the best solution would be using the exact same technology stack and tools for all the clouds your workloads are running on.

You need to rely on a partner and solution portfolio that could address or solve anything (or almost anything) you are building in your IT landscape. As I already said, VCF and Tanzu are just a few pieces of the big puzzle. Important would be an end-to-end approach from any layer or perspective.

Therefore, I believe, VMware is very relevant and very well-positioned to support your journey to the multi-cloud.

The application you migrate or modernize need to be accessed by your users in a simple and secure way. This would lead us for example to the next topic, where we could start a discussion about the digital workspace or end-user computing (EUC).

Talking about EUC and the future-ready workplace would involve other IT initiatives like hybrid or multi-cloud, application modernization, data center and cloud networking, workspace security, network security and so on. A discussion which would touch all strategic pillars VMware defined and presented since VMworld 2019.

VMware 5 Strategic Pillars

If your goal is also to remove silos, provide a better user and admin experience, and this in a secure way over any cloud, then I would say that VMware’s unique platform approach is the best option you’ll find on the market.

And since VMware can and will co-exist with the hyperscalers, and even run on top of all them, I would consider to talk about the “big four” and not “big three” hyperscalers from now on.

Cross-Cloud Mobility with VMware HCX

Cross-Cloud Mobility with VMware HCX

Update 10th Septemeber 2020: vSphere 7.0 (VDS 7.0) and NSX-T 3.0.1+ are supported since the HCX R143 release which has been made available on September 8, 2020

https://docs.vmware.com/en/VMware-HCX/services/rn/VMware-HCX-Release-Notes.html 

Most people think that VMware HCX is a only migration tool that helps you moving workloads to a vSphere based cloud like VMware Cloud on AWS, Azure VMware Solution or Google Cloud VMware Engine. But it can do so much more for you than only application or workload migrations. HCX is also designed for workload rebalancing and business continuity across data centers or VMware clouds. Why I say “across VMware clouds” and not only “clouds”?

A few years ago everyone thought or said that customers will move all their workloads to the public cloud and the majority of them don’t need local data centers anymore. But we all know that this perception has changed and that the future cloud operation is model hybrid.

A hybrid cloud environment leverages both the private and public clouds to operate. A multi-cloud environment includes two or more public cloud vendors which provide cloud-based services to a business that may or may not have a private cloud. A hybrid cloud environment might also be a multi-cloud environment.

We all know that the past perception was an illusion and we didn’t have a clue where the hyperscalers like AWS, Azure or GCP would be in the next 5 or 7 years. And I believe that even the AWS and Microsoft didn’t expect what is going to happen since we observed interesting shifts in the last few years.

Amazon Web Services (AWS) has been launched 14 years ago (2006) to provide web services from the cloud. At AWS re:Invent 2018 the CEO Andy Jassy announced AWS Outposts because their customers have been asking for an AWS option on-premises. In the end, Outpost is just an extension of an AWS region into the own data center, where you can launch EC2 instances or Amazon EBS volumes locally. AWS already had some hybrid services available (like Storage Gateway) but here we talk about infrastructure and making your own data center part of the AWS Global Infrastructure.

Microsoft Azure was released in 2010 and the first technical preview of Azure Stack has been announced in 2016. So, Microsoft also realized that the future cloud model is a hybrid approach “that provides consistency across private, hosted and public clouds”.

Google Cloud Platform (GCP) offers cloud computing services since 2008. Eleven years later (in 2019) Google introduced Anthos that you can “run an app anywhere –  simply, flexibly and securely”.

All the hyperscalers changed their cloud model to provide customers a consistent infrastructure with consistent operations and management as we understand now.

VMware is coming from the other end of this hybrid world and has the same overall goal or vision to make a hybrid or multi-cloud reality. But with one very important difference. VMware helps you to abstract the complexity of a hybrid environment and gives you the choice to run your workloads in any cloud infrastructure with a cloud-agnostic approach.

As organizations try to migrate their workloads to the public, they face multiple challenges and barriers:

  • How can I migrate my workload to the public cloud?
  • How long does it take to migrate?
  • What about application downtime?
  • Which migration options do I have?
  • Which cloud is the best destination for which workloads?
  • Do I need to refactor or develop some applications?
  • Can I do a lift and shift migration and modernize the application later?
  • How can I consistently deploy workloads and services for my multi-cloud?
  • How can I operate and monitor (visibility and observability) all the different clouds?
  • What if tomorrow one the public cloud provider is not strategic anymore? How can I move my workloads away?
  • How can I control costs over all clouds?
  • How can I maintain security?
  • What about the current tools and 3rd party software I am using now?
  • What if I want to migrate VMs back from the public cloud?
  • What if I want to move away/back somewhen from a specific cloud provider?

In summary, the challenges with a hybrid cloud are about costs, complexity, tooling and skills. Each public cloud added to your current on-premises infrastructure is in fact a new silo. If you have the extra money and time and don’t need consistent infrastructures and consistent operations and management, you’ll accept the fact that you have a heterogeneous environment with different technology formats, skill/tool sets, operational inconsistencies and security controls.

If you are interested in a more consistent platform then you should build a more unified hybrid cloud. Unified means that you provide consistent operations with the existing skills and tools (e.g. vCenter, vRealize Automation, vRealize Operations) and the same policies and security configuration over all clouds – your data center, public cloud or at the edge.

To provide such a cloud agnostic platform you need to abstract the technology format and infrastructure in the public cloud. This is why VMware built the VMware Cloud Foundation (VCF) platform that delivers a set of software-defined services for compute, storage, networking, security and cloud management for any cloud.

VMC on AWS, Azure VMware Solution, Google Cloud VMware Engine and all the other hybrid cloud offerings (IBM, Oracle, Alibaba, VCPP) are based on VMware Cloud Foundation. This is the underlying technology stack you would need if your goal is to be independent and to achieve workload mobility between clouds. With this important basic understanding we can take a closer look at VMware HCX.

VMware HCX Use Cases

HCX provides an any-to-any vSphere workload mobility service without requiring retrofit as we use the same technology stack in any cloud. 

VMware HCX Use Cases

HCX enables you to schedule application migrations of hundreds or thousands of vSphere VMs within your data centers and across public clouds without requiring a reboot or a downtime.

If you would like to change the current platform or have to modernize your current data center, HCX allows you to migrate workloads from vSphere 5.x and non-vSphere (Hyper-V and KVM) environments.

VMware HCX Migration

Workload rebalancing means providing a mobility platform across cloud regions and cloud providers to move applications and workloads at any time for any reason.

Workload mobility is cool and may be the future but is not possible today as the public cloud’s egress costs would be way too high at the moment. Let’s say you pay $0.05 per GB when you move data away from the public cloud to any external destination, this would generate costs of $2.50 for a 50GB virtual machine.

Not that much, right? If you move away 500 VMs, the bill would list $1’250 for egress costs. Evacuating VMs from one public cloud to another one is not so expensive if it happens three or four times a year. But if the rebalancing should happen at a higher cadence, the egress costs would get too high. But we can assume that this fact will change in the future as the public cloud computing prices will come down in the future. 

HCX Components and Services

HCX is available with an Advanced and Enterprise license. The Advanced license services are standard with HCX and are also included in the HCX Enterprise license. The Enterprise license is needed when you migrate non-vSphere workloads into a vSphere environment. This capability is called OS Assisted Migration (OSAM).

HCX Services

The HCX Advanced features are included in a NSX Data Center Enterprise Plus license. With a managed service like VMware Cloud on AWS or Azure VMware Solution HCX Advanced is already be included.

HCX Connector Advanced License

If you want to move workloads from a vSphere environment to a vSphere enabled public cloud, you don’t need the complete VMware Cloud Foundation stack at the source site:

  • On-premises vSphere version 5.5 and above
  • Minimum of 100 Mbps bandwidth between source and destination
  • Virtual switch based on vDS (vSphere Distributed Switch), Cisco Nexus 1000v or vSphere Standard Switch
  • Minimum of virtual machine hardware version 9
  • VMs with hard disks not larger than 2TB

Depending on the HCX license and services you need, you have to deploy some or all of the HCX components. HCX comprises components and appliances at the source and destination sites.

HCX Manager Destination

The HCX Connector services and appliances are deployed at the destination site first before you are going to deploy the virtual appliances at the source site (HCX Interconnect appliance).

HCX Interconnect Appliance Download Link

After you deployed the appliances at the source site, you can create the site pairing.

HCX Site Pairing

As soon as you have installed HCX in both sites, you can manage and configure the services within the vSphere Client.

HCX in vSphere Client

After a successful site pairing, you can start to create the HCX Service Mesh.

The Multi-Site Service mesh is used to create a secure optimized transport fabric between any two sites managed by HCX. When HCX Migration, Disaster recovery, Network Extension, and WAN Optimization services are enabled, HCX deploys Virtual Appliances in the source site and corresponding “peer” virtual appliances on the destination site. The Multi-Site Service Mesh enables the configuration, deployment, and serviceability of these Interconnect virtual appliance pairs.

HCX Service Mesh

In the HCX site-to-site architecture, there is notion of an HCX source and HCX destination environment. Depending on the environment, there is a specific HCX installer:

HCX Connector (previously HCX Enterprise) or HCX Cloud. HCX Connector is always deployed as the source. HCX Cloud is typically deployed as the destination, but it can be used as the source in cloud-to-cloud deployments. In HCX-enabled public clouds, the cloud provider deploys HCX Cloud. The public cloud tenant deploys HCX Connector on-premises.
The source and destination sites are paired together for HCX operations. 

In both the source and destination environments, HCX is deployed to the management zone, next to each site’s vCenter Server, which provides a single plane (HCX Manager) for administering VMware HCX. This HCX Manager provides a framework for deploying HCX service virtual machines across both the source and destination sites. VMware HCX administrators are authenticated, and each task authorized through the existing vSphere SSO identity sources. VMware HCX mobility, extension, protection actions can be initiated from the HCX User Interface or from within the vCenter Server Navigator screen’s context menus.

In the NSX Data Center Enterprise Plus (HCX for Private to Private deployments), the tenant deploys both source and destination HCX Managers.

The HCX-IX service appliance provides replication and vMotion-based migration capabilities over the Internet and private lines to the destination site whereas providing strong encryption, traffic engineering, and virtual machine mobility.

The VMware HCX WAN Optimization service appliance improves performance characteristics of the private lines or Internet paths by applying WAN optimization techniques like the data de-duplication and line conditioning. It makes performance closer to a LAN environment. It accelerates on-boarding to the destination site using Internet/VPN- without waiting for Direct Connect/MPLS circuits.

The VMware HCX Network Extension service appliance provides a late Performance (4–6 Gbps) Layer 2 extension capability. The extension service permits keeping the same IP and MAC addresses during a Virtual Machine migration. Network Extension with Proximity Routing provides the optimal ingress and egress connectivity for virtual machines at the destination site.

 

Using VMware HCX OS Assisted Migration (OSAM), you can migrate guest (non-vSphere) virtual machines from on-premise data centers to the cloud. The OSAM service has several components: the HCX Sentinel software that is installed on each virtual machine to be migrated, a Sentinel Gateway (SGW) appliance for connecting and forwarding guest workloads in the source environment, and a Sentinel Data Receiver (SDR) in the destination environment.

The HCX Sentinel Data Receiver (SDR) appliance works with the HCX Sentinel Gateway appliance to receive, manage, and monitor data replication operations at the destination environment.

HCX Migration Types

VMs can be moved from one HCX-enabled data center using different migration technologies or types.

HCX Migration Types

HCX cold migration uses the VMware NFC (Network File Copy) protocol and is automatically selected when the source VM is powered off.

HCX vMotion

  • This option is designed for moving a single virtual machine at a time
  • There is no service interruption during the HCX vMotion migration
  • Encrypted vMotion between legacy source and SDDC target
  • Bi-directional (Cross-CPU family compatibility without cluster EVC)
  • In-flight Optimization (deduplication/compression)
  • Compatible from vSphere 5.5+ environments (VM HW v9)

HCX Bulk Migration

  • Bulk migration uses the host-based replication (HBR) to move a virtual machine between HCX data centers
  • This option is designed for moving virtual machines in parallel (migration in waves)
  • This migration type can set to complete on a predefined schedule
  • The virtual machine runs at the source site until the failover begins. The service interruption with the bulk migration is equivalent to a reboot
  • Encrypted Replication migration between legacy source and SDDC target
  • Bi-directional (Cross-CPU family compatibility)
  • WAN Optimized (deduplication/compression)
  • VMware Tools and VM Hardware can be upgraded to the latest at the target.

HCX Replication Assisted vMotion

VMware HCX Replication Assisted vMotion (RAV) combines advantages from VMware HCX Bulk Migration (parallel operations, resiliency, and scheduling) with VMware HCX vMotion (zero downtime virtual machine state migration).

HCX OS Assisted Migration

This migration method provides for the bulk migration of guest (non-vSphere) virtual machines using OS Assisted Migration to VMware vSphere on-premise or cloud-based data centers. Enabling this service requires additional HCX licensing.

 

  • Utilizes OS assisted replication to migrate (conceptually similar to vSphere replication)
  • Source VM remains online during replication
  • Quiesce the source VM for final sync before migration
  • Source VM is powered off and the migrated VM is powered on in target site, for low downtime switchover
  • VMware tools is installed on the migrated VM

Cross-Cloud Mobility

Most customers will probably start with one public cloud first, e.g. VMC on AWS, to evaluate the hybridity and mobility HCX delivers. Cross-cloud monility is maybe not a requirement today or tomorrow but gets more important when your company has a multi-cloud strategy which becomes reality very soon.

If you want to be able to move workloads seamlessly between clouds, extend networks and protect workloads the same way across any cloud, then you should consider a VMware platform and use HCX.

HCX Cross-Cloud Mobility

Let’s take network and security as an example. How would you configure and manage all the different network, security, firewall policies etc. in your different clouds with the different infrastructure and security management consoles?

If you abstract the hyperscaler’s global infrastructure and put VMware on top, you could in this case use NSX (software-defined networking) everywhere. And because all the different policies are tied to a virtual machine, it doesn’t matter anymore if you migrate a VM from host to host or from cloud to cloud.

This is what you would call consistent operations and management which is enabled by a consistent infrastructure (across any cloud).

And how would you migrate workloads in a very cost and time efficient way without a layer 2 stretch? You would have to take care of re-IPing workloads and this involves a lot of changes and dependencies. If you have hundreds of applications then the cloud migration would be a never ending project with costs you could not justify.

In the case you need to move workload back to your own on-premises data center, HCX also gives you this advantage.

You have the choice in which cloud you want to run your applications, at any time.

 

HCX and vSphere 7

At the time of writing HCX has no official support for vSphere 7.0 yet. I tested it in my home lab and ran into an error while creating the Service Mesh. At least one other colleague had the same issue with vSphere 7 using NSX-T 3.0 and VDS 7.0.

HCX vSphere 7 Error

I would like to thank Danny Stettler for reviewing and contributing. 🙂 Big kudos to you, Danny! 🙂

I hope the article has helped you to get an overview what HCX and a hybrid cloud model really mean. Drop a comment and share your view and experience when it comes to cloud strategies and migrations.

 

Know Your Options with Citrix and VMware

Know Your Options with Citrix and VMware

No, this is not an article about Citrix vs. Horizon and which product is better. And I think that you should not compare Citrix and VMware anymore. If you are still reading and haven’t closed the tab in your browser yet, you made the right decision. The intent of this article is to help you better understand when the usage of Citrix Virtual Apps and Desktops (CVAD) makes sense, which VMware products could complement a CVAD infrastructure and the different options you have with VMware Horizon.

I think it is a very big plus that I worked for Citrix before and still have some technical knowledge. This gives me more credibility in front of the customer and I am not just someone from a vendor, who tries to blame or downplay the other competitor to sell his on stuff. In fact, I always tell my customers how good Citrix is – there is no doubt about that.

But people are still stuck in the past and have the knowledge from four or six years ago. VMware Horizon has evolved into a very mature virtual apps and desktops solution and at the same time VMware’s products evolved as well and the story and product portfolio are better than ever.

Would have asked me a few years ago, no matter if I would be still with Citrix or already with VMware, VMware Horizon had some serious (feature) gaps and differences (e.g. display protocol) compared to Citrix. But Horizon has transformed into a equal player in the market and can do almost the same as CVAD (formerly XenDesktop and XenApp).

Note: I’m not saying that VMware Horizon has reached feature parity compared to Citrix

Let’s see which enhancements or new features have been released in the last 18 months for Horizon:

  • A lot of enhancements and closed feature gaps for the Horizon HTML5 console (now default)
  • RDS Drain Mode and RDSH Load Balancing configurable from UI
  • Improved CDR (Client Drive Redirection) performance
  • Increased CPA (Cloud Pod Architecture) scale up to 250k sessions
  • Session “pre-launch”
  • Two-Factor Re-Authentication
  • Client UI redesign
  • vGPU vMotion (came with vSphere 6.7 U1)
  • VM hosted apps (published applications from Win10 desktop pools)
  • Longer Lived Instant Clones
  • Horizon Cloud Services Enhancements & WVD support for Horizon Cloud on Azure
  • VMware Skyline Log Assist
  • App Volumes 4
  • New REST APIs
  • Bandwidth savings in Blast (with Blast Codec)
  • CPU utilization by Blast has been reduced
  • Blast Extreme HEVC High Color Accuracy support
  • Automatic codec switching based on screen content
  • NSX Advanced Load Balancer (Avi LB) support

As you can see, a lot work has been done and a lot of time has been invested to make Horizon better! These improvements are one of many why I think it’s useless to compare Citrix vs. Horizon, because both can basically do the same if you ask me.

Note: Horizon 8.0 is coming very soon and the beta program for it starts in a few weeks! Stay tuned for more enhancements and innovation. 🙂

Citrix and VMware – Four Options

When I think about Citrix and VMware, there are four options which come up in my mind how a customer could move forward at any given time:

  1. Replace Citrix with Horizon
  2. Integrate Citrix with Workspace ONE
  3. Enhance Citrix with Horizon or Workspace ONE components
  4. Enhance Citrix with other VMware components
  5. Use Citrix and VMware Horizon (yes, there are customers with both!)

Replace Citrix with Horizon

The first option is the most obvious one and can happen from time to time due to various reasons. Sometimes the customer is just not happy anymore (technical or commercial) or wants to try something new because of one or more of the other listed options (integration and enhancements in place already).

A migration would be very easy on paper. StoreFront could be replaced by Workspace ONE Access (formerly vIDM), the VDA installed on RDS hosts or virtual desktops need to be replaced with the Horizon agent and on the client side the Citrix Workspace App (Citrix Receiver) gets replaced by a Horizon Client (including HTML5 client).

Caution: Even if it’s technically possible to uninstall Citrix Virtual Desktop Agenda (VDA) and install the Horizon Agent after, this is not something a good consultant would recommend normally. Do it right and rebuild a clean image and test it before going in production. 

VMware and Citrix Partnership

A replacement could also be done in parallel where you install a Horizon infrastructure beside the current Citrix environment and move the users over whenever you are ready.

If you are running your desktops on Azure together with Citrix Cloud, then the Citrix Cloud piece can be replaced with the Horizon Cloud Service on Azure. Citrix and VMware Horizon are both supported if you are looking for a connection broker for your Windows Virtual Desktops (WVD).

Integrate Citrix with Workspace ONE

The second option doesn’t come up very often. If a Citrix customer is using CVAD only and no Citrix Endpoint Management (formerly known as XenMobile) or Microsoft Intune (or MobileIron) and is considering Workspace ONE for their unified endpoint management of iOS, Android, macOS or Windows 10 clients, then mutual customers could use Workspace ONE (WS1) Access as the web portal or application catalog and single point of access for any application.

As just mentioned already, Workspace ONE users and devices access Citrix-published resources by integrating their Citrix deployment with Workspace ONE Access, which offers an application portal, single-sign on capabilities, conditional access and many other features. Citrix-published resources include applications and desktops from any CVAD infrastructure starting from XenApp 6.0.

All entitlements are still configured in Citrix Studio and you just have to sync these users and groups to the WS1 Access services from Active Directory first.

Beside WS1 Access you need one additional component called the Integration Broker, which can be installed on a Windows Server. The Integration Broker is responsible for the communication with all Citrix farms/sites. The WS1 Access connectors then communicate with the Integration Broker.

Workspace ONE Integration Broker

More information can be found here. That’s all what is needed for the integration with Workspace ONE.

Enhance Citrix with Horizon or Workspace ONE components

VMware has customers with a large Citrix footprint of several thousand users. And some of these customers are using Horizon components together with their Citrix infrastructure. The two most used Horizon components in a Citrix infrastructure are:

I am not up to date anymore what Citrix App Layering, Profile Management (UPM) and Workspace Environment Management (WEM) can do for you today. But App Volumes would replace App Layering and Dynamic User Environment (DEM) would replace UPM and WEM in a Citrix environment.

Don’t know if this still is the case, but a few years ago App Layering had very limited features, didn’t perform and the handling of layers was a pain. And WEM just didn’t scale in larger Citrix environment. Probably Citrix UPM still is doing its awesome job but is leveraging FSLogix for profile and O365 container management and I assume that WEM is also installed more nowadays.

If Citrix App Layering is in use, then probably the FSLogix Application Masking feature could be used as well to hide some components in the image, which also allows the admin to manage fewer golden images. This is something you also can do with Dynamic Environment Manager in combination with App Volumes.

Before FSLogix was available to almost every joint Citrix/VMware and Microsoft customer, it totally made sense to use something like DEM for the user environment management, as DEM has similar features as FSLogix.

To understand the integration of FSLogix and AV and DEM better, this article from VMware’s Digital Workspace TechZone is for you. 

Maybe you ask yourself now how you could get App Volumes and Dynamic Environment Manager for your Citrix environment? Well, there are a few ways and options:

  • Buy the “Horizon Enterprise” or “Horizon Apps Advanced” edition which includes AV and DEM (yes, can happen)
  • Buy the “Workspace ONE Enterprise” edition which includes “Horizon Apps Advanced”
  • Buy the “Workspace ONE Enterprise for VDI” edition which includes “Horizon Enterprise”

You have to buy another license from another vendor, yes. But, let me explain why this could make sense.

Scenario 1 – Citrix customer is buying Workspace ONE Enterprise

Let’s assume you are a Citrix customer and use CVAD to publish applications to your users, but want to manage your iOS, Android, macOS and Windows 10, IoT devices with one solution or platform. That’s the moment when you go for Workspace ONE as your Unified Endpoint Management (UEM) platform. Here’s what you get with Workspace ONE Enterprise:

  • iOS, Android, macOS, Windows 10 and IoT device management (MDM/UEM)
  • Workspace ONE Access
  • Application delivery and management (mobile and desktop)
  • Mobile SSO
  • Workspace ONE productivity apps (email, tasks, notes, content/file repository, web, card scanner)
  • Multi-Factor Authentication (MFA) with “Workspace ONE Verify” mobile application
  • Workspace ONE Intelligence (SaaS-based intelligence and automation engine including reporting)
  • Add-on: Remote Management of any device based on Workspace ONE Assist
  • Add-on: Workspace Security (Carbon Black offerings)
  • Horizon Apps Advanced

The Horizon Apps Advanced edition includes the following:

  • RDS published apps (no desktop OS, only server OS) and session-based desktops
  • ThinApp (not included with WS1 Enterprise)
  • App Volumes
  • Dynamic Environment Management
  • vSphere Desktop

As you can see, you are removing silos in your digital workspace and can use App Volumes and Dynamic Environment Management at the same time to enhance your Citrix infrastructure.

Scenario 2 – Citrix customer is buying Workspace ONE Enterprise for VDI

The difference between scenario 1 and scenario 2 is the Workspace ONE Enterprise for VDI license, which includes the following components:

  • Published desktops and apps (server OS and desktop OS incl. Linux)
  • App Volumes
  • ThinApp (not included with WS1 Enterprise for VDI)
  • Dynamic Environment Management
  • vRealize Operations for Horizon (not included with WS1 Enterprise for VDI)
  • vSphere Desktop
  • vSAN Advanced for Desktop with All-Flash

WS1 Enterprise for VDI makes it possible to have VDI based on the Windows desktop operating system (e.g. Windows 10) as well and adds the infrastructure capability to run your desktop workloads on vSAN enabled clusters! The only thing which differs from the regular standalone Horizon editions, is, that ThinApp and vRealize Operations are not part of the suite. If you have a lot of legacy apps or you need application virtualization or isolation, then take a look at ThinApp.

Applications installers such as MSI files can be packaged into a portable EXE file and can then be run on any physical or virtual Windows PC and delivered with App Volumes (RDS/VDI) or with Workspace ONE (persistent VDI desktop or physical desktop).

And you get the “vSphere for Desktop” edition in both cases which is another killer argument why you could buy Workspace ONE Enterprise (for VDI) licenses as a Citrix customer.

vSphere Desktop

I don’t have any confirmed number, but I assume that 70% of the Citrix customers are using VMware vSphere as their hypervisor. Each regular Horizon edition has vSphere Desktop included which many people are not aware of.

vSphere for Desktop is a special edition, which provides the full range of features of the vSphere Enterprise Plus edition:

  • The new image management feature to patch, update or upgrade ESXi clusters (vSphere 7.0)
  • vCenter Server profiles and update planner (vSphere 7.0)
  • Distributed vSwitch
  • Secure access and account management with ADFS (vSphere 7.0)
  • Distributed Resource Scheduler (DRS)
  • Storage DRS
  • Nvidia GRID vGPU

vSphere Desktop is licensed based on the total number of powered-on VMs and has no processor limitation. It’s available in a pack size of 100 desktop VMs with up to 100 users per pack. VERY IMPORTANT: vSphere Desktop can be used for a VDI environment only and a vCenter license is not included in vSphere for Desktop.

This is the only restriction mentioned in the vSphere Desktop FAQ:

vSphere Desktop can be used only to host a desktop
virtualization environment or desktop management and
monitoring tools. Each pack of 100 VMs can be used for
up to 100 users. You can use vSphere Desktop for desktop
management and monitoring tools in a VDI environment
only. Desktop licenses covered by this provision, however,
may not be managed by the same instance of VMware
vCenter that is being used to manage non-desktop
OS virtual machines.

So, what is considered as a “desktop virtualization environment” including monitoring tools? Normally you would separate your Citrix or Horizon infrastructure servers from the virtual machines which provide the virtual desktops and applications. But this design is more a leading practice and recommended by reference architectures and therefore it is technically possible to mix the RDS and VDI virtual machines with the infrastructure servers like:

  • Connection Server / Delivery Controller
  • Workspace ONE Access / StoreFront
  • Unified Access Gateway / NetScaler
  • Active Directory
  • Monitoring Tools (vRealize Operations / Director)
  • any “other infrastructure directly related to and exclusive to the VDI environment”

In a Citrix Virtual Apps and Desktop environment you can use vSphere Desktop to provide the virtual machines (desktops) and the underlying infrastructure. In this use case, you are licensed per virtual machine and virtual machines used to host the infrastructure servers. These two numbers will be counted against your “total powered-on VM” count. If your Citrix environment has a 100-pack of vSphere Desktop licenses and you host 85 VDI desktops and 15 VMs that host the Citrix VDI environment, then you have used up all the 100 vSphere Desktop licenses.

vSAN Advanced for Desktop

vSAN Advanced for Desktop is shipped together with Horizon Advanced, Horizon Enterprise and Workspace ONE Enterprise for VDI. This license is available for customers using vSAN exclusively for a VDI infrastructure.

Horizon Universal License

The Horizon Universal License is a single subscription-based license, which is included in the Workspace ONE Enterprise edition and serves as an entitlement for all Horizon products, namely Horizon Cloud (including Horizon Cloud Apps) and Horizon on-premises (including Horizon Apps). Thus, the universal license entitles you for the following solutions:

This universal license gives customers the choice to start with an on-premises Horizon deployment and to move to the cloud (or vice versa) without requiring a new license.

Note: Because it’s the universal license and not a regular Horizon license, which is included in the WS1 editions, vRealize Operations (vROps) is not part of this subscription bundle. If needed, vROps can be bought as a standalone license.

Thin Client Management

I thought it is worth mention it here. Keep in mind that you could use a platform like Workspace ONE to manage your thin clients. If your environment is heavily using thin clients you could “build” your own thin client based on Windows 10 IoT Enterprise and manage it via Workspace ONE.

E.g. Workspace ONE can manage Dell Wyse 5070 thin clients with Windows 10 IoT Enterprise. If needed, WS1 can configure the Unified Write Filter (UWF) feature to protect your thin client drives for any changes (saved data, setting changes or app installations). This is also helpful for increasing security for kiosk PCs in hotels, public spots, internet cafés etc. or for devices where it’s not expected to have new application frequently added.

WS1 Unified Write Filter

Enhance Citrix with other VMware components

We know that you could make your Citrix environment “better” with Horizon components like App Volumes or Dynamic Environment Manager and vSphere components like vSphere and vSAN. But there are other products and components which could make sense in a Citrix environment.

I believe, today, VMware has something which you could call a partnership and both CTOs are clearly leading the way:

Citrix Partnership VMware

 

I don’t know if it ever happened before that Citrix mentioned VMware on stage at Synergy, but the announcement from the above picture brings me to my first solution which you could use for your Citrix deployment.

VMware Cloud on AWS

What has been announced at Citrix Synergy 2019? The intent to officially support CVAD running on VMware-based clouds, starting with VMware Cloud on AWS. Many organizations are evaluating or even using a hybrid cloud approach already. This announcement should help Citrix customers, who are running their workloads on vSphere already, to seamlessly move to the cloud to experience a consistent infrastructure with consistent operations.

Because you are using the same technology stack on-prem and in the cloud, this allows you to easily bring your RDS and VDI golden images to the cloud without any a conversion.

I see two deployments options here. Either you leverage the Citrix Cloud services (use VMC as a resource location) or manually install your Citrix infrastructure like you would normally do in your on-premises environment.

VMC on AWS is Citrix-Ready

Note: VMC on AWS is citrix-ready since Q4 2018!

CVAD on VMC on AWS

If you would like to know more about running Citrix Virtual Apps and Desktops with VMC on AWS, please watch the VMworld 2019 recording of the session “Building Global Citrix Virtual Apps and Desktops with VMware Cloud on AWS (HBI2247BU)“. There’s also a recording of the US 2019 session “Building Global Citrix Virtual Apps and Desktops with VMware Cloud on AWS“, presented by Andrew Morgan and James Hsu.

Interesting facts:

  • It takes about 60-70min in average to deploy a new SDDC on VMC on AWS
  • 12min is the average time to add a new host
  • Stretched clusters give you a guaranteed SLA of 99.99%
  • Sync your VM templates with your Content Library
  • Andrew and James deployed 100 Win10 desktops in 5min only
  • PVS and MCS both work on VMC on AWS

NSX – Software-Defined Networking

Digital transformations are nothing new, but get more complex with newer technologies we have today. One very important topic which came up in 2019 and is one of the most important trends for 2020 is “cyber security” or “zero trust security”. VMware and Citrix are both pointing to a zero trust approach to protect the workforce, any app and data. VMware has defined 5 pillars of zero trust for a digital workspace and “transport/session trust” is one of them with these parameters:

  • Micro-Segmentation
  • Transport Encryption
  • Session Protection

For secure transport of a user’s session you would use appliances like the Unified Access Gateway (UAG) or Citrix NetScaler. To achieve a trusted network access within the data center and between workloads, you’ll need something like NSX and micro-segmentation. Citrix has only a SD-WAN solution to protect branch offices and branch users, but no solution for micro-segmentation. What is micro-segmentation and why is it important?

Imagine that network policies can be bound to a virtual machine or in our case to a virtual desktop and dynamically follow a virtual desktop. This is very helpful in the case of VMC on AWS for example. You can easily move the workload to the cloud and move the networking policies together with the VM, because the underlying stack on VMC on AWS (based on VMware Cloud Foundation) includes NSX and the vSphere hypervisor.

How would you secure the communication and access between desktops in the same VLAN? All desktops on a VLAN can communicate freely and one compromised desktop allows lateral movement. With NSX we can provide granular control of desktops and user/group based access control. This is micro-segmentation.

NSX Micro-Segmentation

Here are two articles about Citrix and NSX from VMware and Citrix:

If you are interested in 100% software-defined networking and are thinking to replace an existing hardware or virtual ADCs (application delivery controllers), take a look at NSX Advanced Load Balancer (formerly Load Balancer from Avi Networks).

NSX Advanced Load Balancer Architecture

Where VMware Horizon differs from Citrix

Now you know the four options you have as a Citrix customer when considering VMware products for your current and future environment. Let me explain you why you shouldn’t compare Citrix and VMware Horizon anymore. To get started, you need to understand all the different options you have and how and where you could consume VMware Horizon:

  • Horizon on-premises
  • Horizon Cloud
  • Horizon DaaS

And with the different desktop virtualization offerings there are also different management responsibilities for the customer, partner and VMware:

VMware Horizon Responisibilities

Customers have the flexibility to choose the level of control they want to have over the Horizon and data center infrastructure. If full control of the solution is needed, then you would probably implement Horizon with vSphere on-premises. For use cases where you only would like to maintain the desktop and apps only without concerning yourself about managing any infrastructure, Horizon Cloud on Azure could be one option.

Horizon On-Premises

The biggest difference for me, if you really want to compare Citrix and VMware in a better way, is to see the big picture. People need to understand that it is totally normal that one vendor sometimes is ahead or behind the competitor. The feature set from both vendors, only considering desktop virtualization, is pretty much the same.

When you start a desktop virtualization project and design the solution, you also have to think about the data center part. I’m am not only talking about Horizon and the storage or network requirements here. It’s important to understand the general strategy and vision of VMware and your employer/customer.

Today, automation is a design requirement and you ideally build your on-premises infrastructure based on public cloud principles. Companies don’t start anymore by buying hardware and think about automation later. They want to buy and build something that can be automated from day 1 like it’s done in the public cloud. Everything needs to be agile and elastic and should be able to change when any kind of change occurs.

Because of that it is essential to understand the cloud infrastructure part very well and this is the big difference between Citrix and VMware. We shouldn’t only talk about EUC (End-User Computing) only, but even consider other projects or domains of the infrastructure:

  • Does it fit in my cloud operating model?
  • Can I use an existing solution to automate it (software and hardware)?
  • How would I move my workloads to the cloud tomorrow?
  • Can I integrate existing solutions in my ecosystem (e.g. security, IPAM etc.)?
  • Can it be integrated in our existing or new platform for modern applications based on containers?
  • What about day 2 operations if I need to expand?
  • Can I reduce my silos and reduce the number of vendors and licenses somehow?

The installation of a complete Horizon (or Citrix) infrastructure can be done in a few days, normally, but larger environments require a lot of automation and integrations into the existing infrastructure. Then we talk about several months and not days or weeks anymore.

Horizon on VMware Cloud Foundation

VMware Cloud Foundation (VCF) is made for any workload and is a hybrid cloud platform which provides a set of software-defined components for compute, storage, networking, security and cloud management. VCF is an engineered solution that integrates the entire VMware stack without the need you dealing with complex interoperability matrixes.

VMware Cloud Foundation Overview

The architecture is built on VMware’s Validated Designs (VVD) to reduce the risk of misconfigurations or design failures. The VCF stack is also used with VMC on AWS or Azure VMware Solutions (AVS) for example. This is another reason that clearly shows that this technology stack is the right for any (VMware) infrastructure. If workload mobility is part of your IT strategy, then only VMware can offer this at the moment.

VCF 4.0 Bill of Materials

VMware Cloud Foundation has a “siloed” approach when it comes to the deployment. Based on different hardware resource pools you can create different so-called workload domains (WLD). Each WLD is a different SDDC instance which is managed by software-defined policies. The Horizon deployment can form one or more VDI WLDs.

VCF WLD Overview

Because it’s a standardized approach, VCF makes it very easy to scale on-demand depending on your needs. To get started you’ll need a management workload domain, which is a special-purpose workload domain dedicated for infrastructure and management components like the SDDC Manager, vCenter Servers, vRealize Suite and NSX. The SDDC Manager is responsible for the creation, update or deletion of a workload domain.

Using the regular standard architecture model for VCF, an environment starts with at least 4 physical servers for the management domain, 3 servers for the VI workload domain (Active Directory, SQL servers, any general infrastructure VM) and 3 servers for a Horizon VDI workload domain. This gives us a starting point of 10 physical servers if you build a complete IT infrastructure from scratch. Otherwise you just need the management domain and VDI workload domain with a total minimum of 7 physical servers.

There is also the option available of a consolidated architecture design for smaller environments. In this design the management and workloads run together on a shared management domain. But the consolidated architecture doesn’t support the automated deployment of Horizon yet.

For the automated deployment of Horizon on VCF you would use the SDDC Manager to deploy Connection Servers, App Volumes, Dynamic Environment Manager and Unified Access Gateways. Let me show you some part of the wizard to create a VDI WLD:

You don’t have to install the components by hand, but still need to do your homework before you can deploy the WLD.

I skipped a few steps. You need to upload the Windows server template, convert an existing VI WLD to a Horizon VDI WLD, configure the Horizon AD service account, provide a SQL server and provide information for the load balancers before you reach the step where you enter the details for the connection servers:

One more App Volumes Manager can be added as well:

If you reached the end, you’ll see a review page to do a final check and after that you can run a validation of all your inputs. The deployment of at least one Connection Server is required, but Horizon Composer Servers, UAGs, App Volumes and DEM are optional components and could be skipped.

To expand a current VDI WLD to install UAGs or just to expand the Horizon Pod (add ESXi hosts or Connection Servers) VCF gives you the option to start small and expand later. In the future it should also be possible to shrink a VDI WLD.

The lifecycle management with VCF is very easy. Available updates for all components are tested for interoperability and then bundled with the necessary logic for the proper installation order. VCF offers automated lifecycle management on a per-cluster basis (one WLD can have one or more clusters). This allows admin to target specific workloads or environments for updates independently of the rest of the environment.

VCF Lifecycle Management

For a VDI workload domain VCF delivers a nice view to see the allocated servers/resources and each component related to this workload domain. 

VCF Horizon Deployment WLD

Horizon on VCF on VxRail

So, we know now that VMware Cloud Foundation is the “easy button” for the deployment of the full vSphere stack including vSAN, NSX, vRealize Operations, vRealize Automation, vRealize Log Insight and so on. VCF on VxRail goes one step further and provides you the “one-click upgrade button” for your vSphere stack including the server hardware and firmware. VxRail bundles are pre-configured and pre-tested and therefore validated by Dell EMC and VMware.

VxRail SDDC Manager

The cool thing with VxRail is, that it gives you flexibility for your workloads and that you can choose between different series based on Dell EMC PowerEdge servers. You have multiple compute, memory, storage, network and graphics (M10, P40, T4) options available to cover your workloads and applications with the right server specifications.

VxRail Server Series

Citrix (on VCF) on VxRail

Since VxRail is an HCI appliance, it can run everything on top. I know some larger Citrix customers who are running their Citrix infrastructure on VxRail. It is also possible to run your Citrix infrastructure on VCF on VxRail on a VI workload domain. The only difference with Horizon is the missing automation and integration into the whole (VCF) stack.

Intrinsic Security

In case you missed it, VMware bought Carbon Black and has a new security business unit now. And this is one very important differentiator in this virtual cloud computing space. If VMware’s software-defined data center is your platform of choice already, it makes sense to use a security solution which can be fully integrated and provided by the same vendor.

VMware Security Solutions with Carbon Black

Imagine, that the endpoint protection agent is already integrated in the Horizon Agent and that you could deliver security from your mobile endpoints (Windows, Mac, Linux) to your workloads (VMs or container) in your data center or any cloud (AWS, Azure, GCP). Sounds too good to be true? No, this where the VMware products are heading, especially with Workspace ONE and Horizon (next-gen AV, behavioral EDR, audit and remediation)! 

Workspace ONE for Horizon

I mentioned it already, Horizon is included in the Workspace ONE Enterprise editions. I haven’t covered the case yet where you could combine Horizon and Workspace ONE. If you provide your users persistent virtual desktops based on Windows 10, then it is also possible to manage those with Workspace ONE as well. This will help if you want to move away from a traditional PC lifecycle management (PCLM) solution and move to a modern management approach. So far this only supported with Horizon on-premises installation. Take a look at the product interoperability matrix:

Workspace ONE for Horizon

For which other use cases could this be useful?

  • Physical desktops with Horizon Agent installed (Remote PC access)
  • Physical servers with Windows 10 installed (e.g. HP Moonshot)

I don’t know if the last option has been tested but Windows 10 is a supported operating system for HP Moonshot cartridges.

Horizon Cloud

The Horizon (Cloud) Service is a group of cloud-based services that deliver features for Horizon deployments. This includes the Windows Virtual Desktop (WVD) on Azure as well since the 17th March 2020. Any customer who is using a Horizon subscription license, such as the universal license, can use the Horizon Service.

Horizon Cloud Service Overview

The goal of Horizon Cloud is to provide a single-pane management UI for the delivery and management of your desktops and applications. This is the overview dashboard which shows some information about the health and capacity of all your Horizon deployments.

Horizon Cloud Dashboard

The Cloud Monitoring Service (CMS), which is one of the central services of the Horizon Service, provides data about the user’s session and issues. It can show you how many users and their user experience are impacted related to issues (latency, protocol, slow logon).

In the administration console you can configure the role-based access (RBAC) for your helpdesk admins. It allows them to log in to the admin console and use the search feature to look up users. The help desk administrator can then look up the user’s sessions and perform troubleshooting or desktop maintenance operations. 

Horizon Cloud Helpdesk

The Image Management Service (IMS) is one of the coolest feature of the Horizon Service. As the name suggests it already, it allows you to manage Horizon images from the cloud. You can create, customize, publish and even version all your different images for your Horizon pods. IMS provides a centralized catalog for your images and these can be automatically replicated across the cloud-connected Horizon pods.

Important note: The current release of Horizon Cloud only supports Windows operating systems and on-premises Horizon pods.

Universal Broker

When I joined VMware in May 2018 I was waiting for a feature like this and tried to explain some product managers (PM) that we need something like the Universal Broker. I was looking for a solution that we can avoid E/W traffic in a Horizon multi-pod deployment. I think I tried to explain it to some of our PMs using Citrix’ Optimal Gateway Routing for
Storefront & NetScaler capability. Nobody understood me, but at least we have it now. 😀

Horizon Universal Broker is the cloud-based brokering technology used to manage and allocate virtual resources from multi-cloud assignments to your end users.

These are the listed key features in the VMware Horizon Cloud Service documentation:

  • Single FQDN for all multi-cloud assignments
  • Global pod connectivity and awareness for optimal performance (no longer need for GSLB and no more E/W traffic)
  • Smart brokering (awareness of geographical sites and pod topology)

This diagram shows the Universal Broker components and how the traffic flow works:

  1. From Horizon Client, the end user requests a virtual desktop by connecting to the Horizon Universal Broker service through the brokering FQDN. The service uses the XML-API protocol to authenticate the Horizon Client user and manage the connection session.
  2. After determining that Pod 1 in Site 1 is the best available source for the desktop, the Horizon Universal Broker service sends a message to the Horizon Universal Broker client, which runs on the Horizon 7 Cloud Connector paired with Pod 1.
  3. The Horizon Universal Broker client forwards the message to the Horizon Universal Broker plugin, which runs on one of the Connection Server instances within Pod 1.
  4. The Horizon Universal Broker plugin identifies the best available desktop to deliver to the end user.
  5. The Horizon Universal Broker service returns a response to Horizon Client which includes the unique FQDN of Pod 1 (typically the FQDN of the Pod 1 load balancer). Horizon Client establishes a connection with the load balancer to request a protocol session with the desktop.
  6. After passing through the local load balancer, the request goes to the Unified Access Gateway for Pod 1. The Unified Access Gateway validates that the request is trusted and prepares the Blast Secure Gateway, PCoIP Secure Gateway, and tunnel server.
  7. The Horizon Client user receives the specified desktop and establishes a session based on the configured secondary protocol (Blast Extreme or PCoIP).

Horizon DaaS

In 2013 VMware acquired Desktone. A company that was specialized in delivering desktops and applications as a cloud service. The product got renamed during the years and kept the name “Horizon DaaS“. This is the reason that Horizon DaaS is not just another version of the classic “Horizon” or “Horizon View” since it was a different product which VMware bought. It’s important to know that there are technical differences/characteristics between Horizon and Horizon DaaS because of this history.

Horizon DaaS is the Horizon Desktop-as-a-Service platform for service providers. Not many people understand and know this specific product and you won’t find a lot of content on blogs about it.

The most recent information, beside the official Horizon DaaS documentation, can now be found here 😉 or on Johan’s blog, where he published a lightboard series about Horizon DaaS.

As a service provider you have different options to provide a “managed desktop” or “DaaS” offering:

  • Dedicated Horizon deployment hosted in your data center (licenses through VCPP rental)
  • Horizon Cloud Service (DaaS offering licensed through VCPP MSP)
  • Horizon DaaS – multi-tenant Horizon deployment hosted in your data center (VCPP rental)

Again, Horizon DaaS should be seen as something different than Horizon, it’s really just not Horizon. But the future strategy and look of the user interface will be aligned with Horizon Cloud, because VMware’s Horizon Cloud Service is powered by Horizon DaaS already.

If multi-tenancy is a key requirement for your business, you’ll have to go with Horizon DaaS. Otherwise the regular Horizon edition or the combination with Horizon Cloud are the right fit. Horizon DaaS and Horizon have common components like vCenter, Agents, UAGs etc., but there are also different appliances with Horizon DaaS which replace components of a regular Horizon deployment.

Horizon DaaS Architecture

With Horizon DaaS you are going to have “Service Provider” appliances, “Tenant” appliances and “Tenant Resource Manager” appliances, which form the DaaS back-end.

The Service Provider Appliance is the first appliance installed in a data center and provides the foundation to install the remainder of the Horizon DaaS application.

The Resource Manager abstracts the specifics about the desktop infrastructure from the tenant appliances and allows multiple Desktop Managers to communicate with their respective virtualization resources. A Resource Manager appliance integrates with the hypervisor and storage infrastructure in a given data center. A single Resource Manager appliance can be shared across multiple tenants.

The Tenant Appliance provides the tenant with both end user and administrative access to their virtual desktops. End users access and manage their individual virtual desktops via the Desktop Portal. Administrators create and manage their virtual desktops via the Enterprise Center. The Tenant Appliance includes the Desktop Manager, a per-tenant resource that manages each tenant’s virtualization resources and communicates with a tenant’s hosts (hypervisors). You associate the desktop manager with a resource manager and one or more host managers.

It’s not 100% clear from the Horizon DaaS 8.0.0 Service Center guide, but a Tenant Appliance replaces the Connection Server you would know from a regular Horizon deployment (one of the differences I was already referring to).

Use what makes sense

For me it is very important that you understand how VMware products can help and that people are aware of all the different options they would have with VMware and Horizon.

You must form your own view and opinion and I hope this article was useful to get facts from both worlds (based on my best knowledge and experience). If you understand Horizon better now, this is already fine for me.

There was no intention to lead the path to a way where you would replace Citrix. The new information should help you to make the right decision for your company, your environment, your needs and use cases. Use the products which make sense for you and make sure you understood all options.