In this article, we will not go into any technical details. Instead, let's understand the terminology of VMware. This short overview will be useful for those who want to understand the difference between the base VMware products. Almost everyone knows about ESXi. What about vSphere and vCenter? People often confuse these terms, but in fact there is nothing complicated about them. Let's break it all down.

Launched in 2001, the VMware ESX hypervisor (formerly known as VMware ESX Server) launched the virtual revolution. Today, VMware is a leading developer of virtualization software products (now part of Dell). Every year and a half, the company releases new software with advanced features that is compatible with large quantity equipment, including SSD drives NVMe, very high capacity hard drives and the latest central processors Intel or AMD.

VMware ESXi

ESXi is a hypervisor; tiny particle software, which is set to physical server and allows you to run multiple operating systems on a single host computer. These operating systems operate separately from each other, but can interact with the outside world through a network. The rest of the computers are connected to the local network (Local Area Network, LAN). Operating systems run on virtual machines (Virtual Machine, VM), each of which has its own virtual hardware.

There are paid and free versions of VMware ESXi. You can order the installation of a free version on . The functionality of the free version is somewhat limited. It allows you to consolidate a limited number of operating systems on one computer, and it cannot be managed through a central management server - vCenter. However, Free ESXi (or VMware ESXi Hypervisor) connects to remote storages where you can create, store and use virtual machines. That is, this remote storage can be shared between multiple ESXi hosts, but not between virtual machines. Virtual machines are "owned" by each host, making central management impossible.

Working with the free version of ESXi is very simple and consists of basic processes: training, testing production processes, checking disaster recovery systems, approving architectural solutions. Using snapshots, you can check the correct operation of Windows patches. Alternatively, this can be useful if you decide to clone your production server using VMware Converter or P2V technology and want to test the Microsoft update package before installing it.

VMware vCenter

VMware vCenter is a centralized management platform for VMware virtual infrastructure. With it, you can manage almost all processes from just one console. The vCenter Server can be installed on Windows or deployed as a pre-configured virtual machine using Photon OS, a powerful Linux-based distribution. VMware used to use the Suse Linux Enterprise Server (SUSE) distribution, but recently switched to Photon OS.

vCenter Server is licensed software. You can buy it in two ways:

• vCenter Server Essentials as part of the vSphere Essentials package. This version of vCenter manages to manage three hosts with two physical processors each. If you have a small company, then you will operate with about 60 VMs, and this version of vCenter will suit you. With the basic set, you get a license not only for vCenter server, but also for ESXi (up to three hosts with two CPUs on each).
• Standalone vCenter Server is a complete standalone vCenter server capable of managing 2,000 hosts with 25,000 working virtual machines. This is a license exclusively for vCenter. The vCenter itself is only part of the licensing puzzle. To manage all hosts from one device, you need a license for each of them. There are three types of licenses: standard, enterprise, enterprise Plus, and each covers one processor. So, if you plan to create a host with two physical processors, then you will need 2 licenses for this one host only.

VMware vSphere

VMware vSphere is the commercial name for the entire VMware suite of products. As mentioned earlier, different software packages cost different amounts of money. The cheapest are the base vSphere essentials or Essentials Plus packages. Is there a difference between them? Yes, but it consists in the number of available functions, and not in the software content itself.

Depending on the type of license, you get access to a certain number of features that can be controlled through the vSphere Web client. There is also a vSphere HTML 5 client program, but it is not yet usable. The company continues to develop it.

The Essentials package does not include High Availability (automatic restart of the VM), vMotion, backup software (VDP), and the ability to use VSAN storage.

The Essentials package is suitable for small businesses that don't need to be online all the time. On the other hand, being able to move your virtual machines to another host and perform host maintenance or upgrades while still online gives you a real advantage. All this can be done during the working day, without interrupting the work of users.

In addition, in the event of an unexpected hardware failure, vSphere High Availability (HA) will automatically restart the virtual machines that failed along with the affected host. These virtual machines are automatically restarted on other hosts from the VMware cluster. It takes a little time for the system to determine which computer has failed and which hosts can temporarily take over its virtual machines. These hosts must have enough memory and CPU power to handle the extra load. As soon as the system finishes the analysis, the VMs are restarted. The whole process is automated and does not require administrator intervention.

Summarize

As you can see, the VMware terminology is pretty easy to understand, as is the difference between ESXi, vSphere, and vCenter. The licensing system is also clear. The hypervisor itself is free, but its functionality is limited, as a result of which such software is not immune from data loss. Therefore, ESXi is intended to be used only in test environments.

The term "virtualization" has become very fashionable lately. The concept of "virtual machine" has ceased to be something exotic and distant. Many organizations, one way or another connected with information technology, have already learned how to use virtual machines in their daily activities to increase the efficiency of their IT infrastructure. The concept of virtualization is now used everywhere and, sometimes, in different contexts: virtualization of storage systems, operating systems, applications. If various aspects of virtualization of hardware resources have been known to specialists for a long time, then the virtualization of operating systems has begun to gain momentum only in the last few years, but at a rapid pace.

So what is virtualization and virtual machines in relation to operating systems? By itself, the term virtualization means the presentation of an object in a user-friendly form, while all implementation details are hidden, and the object itself has the usual interfaces for interacting with the external environment in relation to it. When talking about the virtualization of operating systems, first of all, they mean the creation of virtual machines - some abstractions that contain their own virtual hardware and software environment that allows you to install and run several instances of operating systems simultaneously on one physical platform. What is it for? First of all, in order to separate the representation of the operating system from the hardware and place several working virtual servers on one physical one with the ability to quickly migrate and restore operating environments. This approach also provides maximum flexibility in terms of server deployment, maintenance, and management.

Most CIOs of large organizations, if not planning partial or full virtualization of their IT infrastructure, are at least seriously thinking about it. In the long term, virtualization projects seem very tempting: increasing the flexibility of the IT infrastructure, increasing the reliability and ensuring high availability of servers, saving on hardware - all these and many other factors attract the heads of IT departments of companies. Nevertheless, many do not realize how much effort it will cost to migrate a physical infrastructure to a virtual one - after all, there are only a few competent specialists in this area now, and the acquisition and deployment of powerful commercial virtualization platforms requires considerable costs. This article will discuss two popular platforms for server virtualization - the free VMware Server and the commercial VMware ESX Server.

Why VMware?

VMware is one of the early players in the emerging virtualization platform market. In 1998, VMware patented its software virtualization techniques and since then has released many effective and professional products for virtualization of various levels: from VMware Workstation aimed at the end user (consumer) to VMware ESX Server designed to host the virtual infrastructure of medium and large enterprises. In a very extensive list of VMware products, you can find many tools to increase the efficiency of the virtualization process, manage virtual servers, and migrate from physical to virtual platforms. In Russia, VMware products are especially popular, since virtualization is still gaining momentum here, and platforms from other vendors, which are not so well known to us, are very “raw” and have much less functionality than VMware counterparts. In addition, VMware virtualization products almost always outperform the competition in various performance tests. And if they talk about the virtualization of Windows operating systems, then these are almost certainly VMware products. Meanwhile, among virtualization platforms, VMware has plenty to choose from:

• VMware Workstation- a platform focused on Desktop-users and intended for use by software developers, as well as professionals in the field of IT,
• VMware Player- a free "player" of virtual machines, designed to run ready-made templates of virtual machines that perform a specific task,
• VMware Server, formerly called VMware GSX Server, aimed at use in the infrastructure of small enterprises to maintain virtual servers
• VMware Ace- a product for creating virtual machines protected by security policies,
• VMware ESX Server- a powerful virtualization platform for medium and large businesses, focused primarily on maintaining a holistic and scalable IT infrastructure,
• VMware Virtual Center- a powerful tool for managing virtualization platforms VMware ESX Server and VMware Server, which has extensive capabilities for server consolidation, configuration and management.
• VMware Fusion is a desktop virtualization product for the Mac platform from Apple.

VMware's formal competitors in terms of server virtualization are Microsoft, Virtual Iron, XenSource, and SWsoft. However, SWsoft's product has a rather narrow scope (hosting), and the developments of other vendors currently look much weaker compared to VMware products.

When to implement virtual infrastructure in a company

First of all, it is necessary to clearly define the goals for which virtualization platforms will be implemented. Then, it is necessary to determine the criteria by which the effectiveness of the implementation of the virtual infrastructure will be evaluated. In addition to purely financial criteria (reduced hardware costs, savings in maintenance), you must also consider improving infrastructure reliability, scalability, flexibility, resiliency, reduced deployment time, downtime, disaster recovery, centralization of management and ensuring high availability. Of course, all these criteria can be expressed in monetary terms, depending on the type and specifics of the environment to be virtualized.

7 reasons to implement server virtualization

• Increasing hardware utilization
  According to statistics, most servers are loaded by 15-20 percent when performing daily tasks. The use of several virtual servers on one physical one will increase it up to 80 percent, while providing significant savings on the purchase of hardware.
• Reduce hardware replacement costs
  Since virtual servers are decoupled from specific hardware, updating the fleet of physical servers does not require re-installation and configuration of the software. The virtual machine can simply be copied to another server.
• Increasing the flexibility of using virtual servers
  In the event that you need to use multiple servers (for example, for testing and working in production) under varying load, virtual servers are the best solution, as they can be painlessly transferred to other platforms when the physical server is under increased load.
• Ensuring high availability
  Backing up virtual machines and restoring them from backups takes much less time and is a simpler procedure. Also, in case of equipment failure, the backup copy of the virtual server can be immediately launched on another physical server.
• Increasing the manageability of the server infrastructure
  There are many virtual infrastructure management products that allow you to centrally manage virtual servers and provide load balancing and live migration.
• Savings on service personnel
  Simplifying the management of virtual servers in the future entails savings on specialists serving the company's infrastructure. If two people can do what four people can do with virtual server management tools, why do you need two extra people making at least $15,000 a year? Nevertheless, it must be taken into account that a lot of money is also needed to train qualified personnel in the field of virtualization.
• Saving on electricity
  For small companies, this factor, of course, does not really matter, but for large data centers, where the cost of maintaining a large fleet of servers includes electricity costs (power, cooling systems), this moment is of considerable importance. The concentration of several virtual servers on one physical will reduce these costs.

When not to implement infrastructure in a company

Despite all the advantages, server virtualization also has some limitations of its application. When planning a virtual infrastructure, consider the following factors:

• Servers have a continuous workload of over 60 percent
  Such servers are not good candidates for virtualization, because when it is migrated to a virtual machine and placed with other virtual servers, it may run out of resources.
• The server uses additional hardware that cannot be virtualized
  Everything is clear here: in the case when the server uses hardware that is not supported by virtualization platform vendors, it makes no sense to virtualize such a server.
• The cost of acquiring and implementing a virtualization platform is too high
  In medium and small organizations, the server infrastructure is not that big, and so are the hardware and maintenance costs. In such a case, you need to be careful about the virtualization procedure, since the purchase of a commercial platform may not be justified.
• Lack of qualified specialists
  In many cases, the process of migrating from physical hardware to virtual machines and further deploying a virtualization platform requires good skills from the people who carry them out. This is especially required when deploying "Bare metal" class platforms (bare metal). If you're not sure that your people have the knowledge to do so, don't start a virtualization project.

VMware Server is a free and powerful virtualization platform for small businesses

The free product VMware Server is a fairly powerful virtualization platform that can be run on servers running Windows and Linux host operating systems. The main purpose of VMware Server is to support small and medium-sized virtual infrastructures of small enterprises. Due to the small complexity of its development and installation, VMware Server can be deployed in the shortest possible time, both on the servers of organizations and on computers of home users.

Previously, this product was distributed under a commercial license and was called VMware GSX Server 3, however, with the growth of opportunities and sales of the powerful VMware ESX Server virtualization platform, VMware did not see any prospects in sales of the VMware Server platform, eventually making the product free. It is worth noting that for this product, VMware relies primarily on sales revenue from Virtual Center for VMware Server, a powerful virtual infrastructure management tool based on VMware Server that provides rich virtual machine interaction and virtual server consolidation capabilities.

Here are the main use cases for the VMware Server product:

• support for several virtual servers on one physical in the company's production
• support for multiple virtual servers for "bundled" testing in the host's virtual network during software development and support
• launching ready-to-use virtual machines (Virtual Appliances) that perform a specific server function
• ensuring high availability of virtual servers (virtual machines can be easily transferred between physical servers)
• creation of backup copies of virtual servers that are easy to restore by obtaining snapshots of the current state of the system (“snapshots”).

VMware Server has a wide range of options for working with virtual machines, including:

• Support for any standard x86 architecture
  VMware Server does not have specific requirements for physical server components - unlike VMware ESX Server, which imposes very specific restrictions on server hardware. Multi-core processors are also supported.
• Support for bidirectional virtual SMP (symmetric multi-processing)
  If the architecture of the physical server allows, virtual machines created in VMware Server can contain two virtual processors, which positively affects the performance of guest systems.
• Support for a large number of host and guest operating systems, a complete list of which is always available on the VMware website
  The number of host operating systems on which VMware Server can be installed is, of course, less than the number of supported guest systems. Meanwhile, almost any known operating system can be installed as a guest. Even if it is not in the list of supported, this does not mean that it cannot be installed.
• Support for 64-bit host and guest operating systems
  As host 64-bit systems, 64-bit operating systems of the family Windows Server 2003, as well as 64-bit versions of Linux systems: Red Hat, SUSE, Mandriva and Ubuntu. The list of supported guest operating systems has also been supplemented with 64-bit versions of Windows Vista, Sun Solaris and FreeBSD.
• Support for IntelVT (Intel Virtualization Technology)
  VMware Server experimentally supports Intel's hardware virtualization technology and allows it to be used to support virtual machines. It is worth noting that according to the research of VMware engineers, hardware virtualization is still slower than software virtualization, so it is not recommended to enable support for software virtualization in order to improve performance.

Acquaintance with VMware Server begins with the virtual machine management console window:

In the workspace of the main program window, you can create a virtual machine, add an existing one, switch the console to another physical host (control the virtual server remotely), and configure host settings.

Creating a virtual machine in VMware Server is a simple and intuitive process and takes just a few steps of the virtual machine creation wizard:

• select the guest system from the list, which will be installed as a guest
• select the name and location of the virtual machine files
• select the type of network interaction between the virtual machine, host OS, other virtual machines and external network
• enter volume virtual disk and click Finish.

After that, if the distribution kit of the installed guest system is on a CD or DVD, just insert it into the drive and press the "Power on" button on the toolbar. If you have the distribution kit of the operating system in the form of an ISO image, select the menu item VM->Settings, go to the CD-ROM element, where we indicate the path to ISO image, click "OK" and "Power On".

The process of installing a guest system in VMware Server is quite simple and does not require a separate description. However, when installing guest OSes, be sure to consider the following points:

• Allocate enough resources to the planned virtual server to perform its functions, but remember that it is always more difficult to increase the number of allocated resources than to reduce it.
• When choosing the type of network interaction for a virtual server, consider security issues, and if the virtual server only needs interaction only in the internal virtual network of the host, do not install Bridged Networking.
• If you are not the only one with access to the virtual server management console, you can make your virtual machine private by checking the VM->Settings->Options->Permissions checkbox.
• Don't forget to install VMware Tools in your guest systems, as installing these add-ons greatly improves the guest experience and performance.
• Try to keep the ratio: no more than 4 virtual machines per physical processor, since more of them will significantly affect the performance of virtual servers.

When maintaining a virtual infrastructure based on VMware Server, you must carefully monitor the load on the server's hardware resources. If any virtual machine does not have enough resources, you need to think about moving it to another server. If you plan to manage a large number of virtual servers, you should consider purchasing the VMware Virtual Center product, which allows you to centrally control many hosts on which VMware Server is installed, group them into clusters, and monitor the load of hosts by virtual machines. To determine the amount of resources to allocate to virtual servers, use the performance counters inside the guest systems. Try to use only virtual SCSI disks because virtual IDEs are slower.

If you need to access the virtual server console from an external network, you can set up a Web client for VMware Server that is integrated with Microsoft IIS. This feature will allow you to control virtual servers over the Internet using a secure SSL (Secure Socket Layer) connection.

VMware Server Specifications

Opportunities	VMware Server 1.0
Ability to run as a service	Yes
Starting virtual machines at the start of the host operating system	Yes
Local control	Thick client, command line
Multiple user access	Yes
Programming Interfaces (APIs)	Yes (C/COM/Perl)
Remote use of the product	Yes (web console)
Remote use of virtual machines	fat client
Managing multiple installations	Not
VM/Core Ratio	2-4
Support for hardware virtualization	Intel VT (Experimental)
Generation of virtual hardware	5
Support for virtual processors inside a virtual machine (VMware Virtual SMP technology)	2 (Experimental)
The maximum amount of RAM allocated for a virtual machine	Up to 3.4 GB
The maximum amount of RAM allocated for all virtual machines	Up to 64 GB
IDE virtual controllers/ disks per virtual machine	1/4
SCSI virtual controllers/disks per virtual machine	4/60
Maximum virtual disk size	Up to 950 GB
Maximum number of virtual network interfaces	4
virtual switches	9
Getting snapshots through the "thick" client	Yes
Getting snapshots via command line	Not
Multiple Snapshots	Not
Cloning virtual machines	Not
Virtual machine groups (Teams)	Not
Debugging virtual machines	Yes
Shared folders service	Not
Drag&Drop support Host Virtual machine	Not
Unsupported host OS	Windows XP Professional (32/64bit) Windows XP Home Windows 2000 Professional Red Hat Linux 7.0 Red Hat Linux 7.1

1. Only on operating systems that support extended memory or have PAE enabled.
2. Available when using VMware Virtual Center.

For the VMware Server product, there are many different commercial and free utilities for managing virtual machines and disks, monitoring performance and maintenance. However, if you're going to take full advantage of VMware Server's capabilities, you should take a closer look at VMware Virtual Center for VMware Server.

VMware ESX Server is an enterprise-class virtualization platform and enterprise virtual infrastructure builder

Among the many virtualization solutions offered by VMware, VMware ESX Server occupies a special place. As the company's flagship development, VMware ESX Server is the basis for building large virtual infrastructures into which other VMware products are integrated.

VMware ESX Server is a "Bare Metal" virtualization platform and is installed on a "clean" server that does not have an operating system and other software. VMware ESX Server is based on the Red Hat Linux operating system, in which VMware engineers have made significant changes, adding many features to support virtualization. This implementation of this platform allows the most efficient use of server hardware resources and provides the best interaction between guest systems and hardware.

VMware ESX Server has many benefits and is the core of a large enterprise virtual infrastructure that requires multiple virtual servers to be maintained, consolidated, highly available, and migrate virtual machines between physical servers. ESX Server is a true concept of building an enterprise IT infrastructure using virtual machines.

In addition to the ESX server, the virtual infrastructure of the enterprise also contains components such as SANs (Storage Area Networks), storage systems (Storages), high-speed connections (Fibre Channel) and virtual networks (VLANs). A virtual infrastructure based on VMware ESX Server allows you to centrally manage all these resources and ensure the smooth operation of virtual servers. The main components of the virtual infrastructure are:

• Own ESX Server.
• VMware VMFS (Virtual Machine File System) file system optimized for virtual machine performance and high availability.
• VMware SMP (Symmetric Multi-Processing)- a technology that allows virtual machines to use all the processors of the host, while having several virtual processors.
• Virtual Infrastructure Client- a powerful tool for remote administration and configuration of VMware ESX Server, running on Windows workstations and having a friendly interface.
• Virtual Center- a tool for centralized management of multiple ESX servers.
• Virtual Infrastructure Web Access- the ability to manage virtual machines from an external network via a secure channel.
• VMware VMotion- a technology that allows you to carry out a "live" migration of a running virtual server to another physical server, without the need to turn off the virtual machine and ensure its uninterrupted operation during migration.
• VMware High Availability (HA)- the ability to automatically restart the necessary virtual machine mirrors on another server in the event of a hardware or software failure of a physical server.
• VMware Distributed Resource Scheduler (DRS)- a component that allows dynamic allocation of resources to virtual machines.
• VMware Consolidated Backup (VCB)- a simple and powerful tool for creating backup copies of virtual machines.
• VMware Infrastructure SDK- a package for developing applications for virtual infrastructure by third-party developers.

When to Use a Virtual Infrastructure Based on VMware ESX Server

• You have a clear idea of ​​how many physical servers you will need to maintain your virtual server infrastructure. Keep in mind that you need to plan the migration of physical servers based on no more than 4-6 virtual machines per physical processor, with an average real load of migrated physical servers of 15%.
• You have selected the appropriate VMware ESX Server edition and virtual infrastructure components, calculated the cost of deploying and maintaining them, and concluded that the implementation is worthwhile because the savings in hardware and maintenance cover these costs.
• You have specialists who will not only competently conduct a virtualization project, but also administrators with sufficient knowledge to maintain and develop a virtual infrastructure every day.
• You are ready to purchase the hardware required by VMware ESX Server. Be careful - the ESX server installation guide says what specific hardware is required to install it. At first glance, everything will seem quite expensive, but if you read the requirements carefully, it becomes clear that a fully functioning ESX Server for training purposes can be assembled for $900.

The procedure for installing VMware ESX Server is quite simple and does not require extensive knowledge from the user. If you followed VMware's recommendations and correctly selected the host server hardware, then installing the ESX server will take no more than an hour. After you install at least one ESX, you will immediately have a lot of questions and problems - this is the price for the opportunities that VMware ESX Server provides. Here are the main recommendations for the first steps after installing ESX:

• To manage the ESX server and create the first virtual machines, use the VMware Infrastructure Client, which can be downloaded from: https://.
• To transfer files to and from ESX, use either WinSCP (slow file transfer, encrypts traffic) or FastSCP (fast file transfer, but it is better not to use it from an external network, since traffic is not encrypted).
• In order to allow the Root user to log in via SSH (Secure Shell), add the line "PermitRootLogon yes" to the "httpd.conf" file on ESX.
• To manage disks, use the "vmkfs-tools" utilities, to monitor the performance of virtual machines - the "esxtop" command, to analyze error reports, use the "var/log/vmware/hostd.log" log.
• Remember, free space on your ESX is always two types of partitions: the first is for the operating system of the ESX itself, the second is vmfs partitions for hosting virtual machines. To find out how much free space is left on both partitions, use the "vdf -h" command.

After you install VMware ESX Server, you will need a tool to migrate from physical to virtual (P2V - Physical to Virtual) servers. VMware offers VMware Converter for this purpose, but you can also use other vendors' solutions. You will constantly have questions - do not hesitate to contact the VMware forums, where its regular visitors will help you in a difficult situation. Ultimately, your ESX will have several virtual machines installed, representing a virtual infrastructure, which is shown in the figure made using the Veeam Reporter program:

The key concepts of virtual infrastructure are: physical adapter (NIC), virtual adapter (vNIC), virtual switch (vSwitch) and virtual network (Vlan). VMware ESX Server allows you to create up to four virtual network adapters for a virtual machine, each of which can be bound to a virtual network created in turn on virtual switches.

A virtual switch is an abstract multiport device that performs circuit switching between virtual networks and virtual network adapters of virtual machines.

A virtual network is an association of several virtual machines into a single network environment in which they interact with each other. If a virtual switch is connected to a physical network adapter, then virtual machines through it will be able to "see" the external, in relation to ESX, network.

In fact, everything is not as complicated as it seems: creating elements of a virtual infrastructure takes very little time and, once configured, such an infrastructure does not require further configuration when new virtual machines are introduced into it.

A little about Virtual Center for VMware ESX Server

As with VMware Server, the use of multiple hosts with VMware ESX Server in an enterprise's IT infrastructure creates the challenge of centrally managing and monitoring the performance of physical hosts. To solve these problems with VMware ESX Server platforms, as well as for VMware Server, the Virtual Center solution is used. Its appearance is shown in the figure:

Virtual Center allows you to monitor multiple hosts on which VMware ESX Server is installed, cluster them and control them using "alarms" - signals about various events. It should be noted that the capabilities of Virtual Center for VMware ESX Server are much greater than those of Virtual Center for VMware Server, which is primarily due to the fact that the VMware ESX Server product itself has much broader capabilities.

What to choose: VMware Server or VMware ESX Server?

From the above, it becomes clear that both products described in the article are aimed at maintaining a virtual server infrastructure and perform similar functions. However, there are significant differences between these two platforms. If a virtual infrastructure based on VMware Server can be built primarily from servers that perform day-to-day tasks in an organization that do not require high availability, speed and flexibility, then VMware ESX Server is a complete platform for supporting an enterprise's vital IT infrastructure in conditions of uninterrupted operation of virtual servers and their support in 24×7×365 mode.

Here are some examples of when to use a VMware Server-based infrastructure:

• support and maintenance of internal servers of the organization,
• performing tasks for testing individual applications,
• simulation of small virtual networks in order to test working server links,
• launching ready-to-use virtual machine templates that act as internal servers of the organization,
• obtaining individual servers ready for rapid migration.

VMware ESX Server should be used for the following tasks:

• streaming application testing in large software development organizations,
• maintenance of external servers of the organization with a high degree of availability, flexibility and manageability,
• modeling large virtual networks,
• reducing the cost of equipment, maintenance and electricity in large organizations and data centers.

Thus, when implementing a virtual infrastructure in an organization, you need to carefully consider the definition of end goals. When deploying a free VMware Server, there will be no special problems with installation and maintenance, there will also be no costs for acquiring the platform itself, however, the organization may lose speed (since virtualization is performed on top of the host operating system), reliability and availability. The introduction of VMware ESX Server will result in great technical difficulties in deployment and maintenance in the absence of qualified specialists. In addition, investing in such a platform may not pay off for small and medium-sized organizations. However, as VMware's experience shows, for large organizations, the implementation of VMware ESX Server, in the long run, turns into serious cost savings.

Taking into account these points will allow you to competently plan the migration of physical servers of your IT infrastructure to virtual ones, while saving not only money, but also time, which, as you know, is also money.

What's new in VMware Cloud Foundation 4?

Recently, we talked about new platform features and other updates to the VMware product line, announced simultaneously with the flagship product. Recall these articles:

Today we will talk about another important update - a new version of the VMware Cloud Foundation 4 hybrid infrastructure solution set. We wrote about the previous version of this VCF 3.9.1 package. As you remember, it is a complex software solution, which includes components from VMware vRealize Suite, VMware vSphere Integrated Containers, VMware Integrated OpenStack, VMware Horizon, NSX, and others running on premium, cloud, or hybrid enterprise infrastructure managed by SDDC Manager.

The fourth version of the VCF includes all the latest components, the articles that we described above:

• vSphere 7
• VMware vSAN 7
• VMware NSX-T
• VMware vRealize Suite 2019
• with Kubernetes support

As we can see, a fundamentally new component has appeared in the VCF stack - VMware Tanzu Kubernetes Grid. We already wrote about the container support infrastructure in the new version of the vSphere platform and. With the new VCF architecture, administrators can deploy and serve applications on Kubernetes clusters using Kubernetes tools and a restful API.

At the same time, vSphere with Kubernetes technology (aka Project Pacific) will provide the following functionality:

• Kubernetes-powered vSphere Pod Services will allow nodes to run directly on the ESXi hypervisor. When an administrator deploys containers through vSphere Pod Services, they receive the same level of security, isolation, and performance guarantees as virtual machines.
• The Registry Service allows developers to store and maintain Docker and OCI images on the Harbor platform.
• Network Services allow developers to manage Virtual Routers, Load Balancers, and Firewall Rules.
• Storage Services allow developers to manage persistent disks for use with containers, Kubernetes clusters, and virtual machines.

All this allows you to get all the benefits of a hybrid infrastructure (VM + containers), which are interestingly described.

Otherwise, VCF 4 acquires all the newest features that the already listed new releases of vSphere, vSAN, NSX-T and others provide.

Separately, it should be noted that vSphere Lifecycle Manager (vLCM) is very tightly integrated with the vSphere 7 platform. vLCM complements the lifecycle management capabilities of virtualization infrastructure components that are already in SDDC Manager, but at a deeper level - namely, at the firmware management level for vSAN hosts ReadyNodes (for example, HBA firmware updates).

Like all other updates of the vSphere line, the VCF 4.0 update is expected in April. Updates can be followed on this page.

Tags: VMware, Cloud, VCF, Update, vCloud, Enterprse

Today we will talk about the Identity Federation services introduced in VMware vSphere 7.

AT modern world Increasingly, corporate infrastructure is moving away from legacy password authentication and moving towards two-factor (2FA) or multi-factor (MFA) authentication practices. The user identification process is always based on 3 key things: something you know (password), something you have (phone) or someone you are (fingerprint).

Identity Federation Services allows you to combine your vCenter Server infrastructure with other Identity Providers such as Active Directory Federation Services (ADFS), in order to unify the process of two- or multi-factor authorization. In other words, users who log in via 2FA to their desktop or cloud service, will use the same procedure for vCenter Server operations.

When connected to one of the authentication providers (for example, ADFS), the vSphere Client will redirect to the login form of this provider when logging in. After authorization on the provider's side, a reverse redirect will be made using a secure token, through which the user will already work with vCenter services.

In terms of user experience, this is similar to, for example, logging into a website using Google or Facebook. The OAUTH2 and OIDC protocols are used to exchange information.

If you enable Identity Federation, you can use traditional Active Directory, Integrated Windows Authentication, and LDAP/LDAPS to authenticate with vCenter Server. However, it must be understood that all these authentication methods do not affect vSphere Single Sign-on (SSO), which is still used to make administrative settings in the vSphere platform itself.

Bob Plankers talks about this mechanism in more detail in the video below:

Tags: VMware, vSphere, Security, Client, Update

Here's what's new in the Ubuntu OVA for Horizon 1.2 image:

• Minimum support for Horizon 7.11 / Horizon Client 5.3 and above later versions
• Minimum support for vSphere 6.7 and later
• Updated OVA template base image on Ubuntu 18.04.4 LTS
• Updated virtual hardware - Virtual Hardware v14
• Added the ability to set a static IP address
• Added USB support 3.0 and USB Redirection (via linux-agent-installer.sh script)
• Added option to select KDE Desktop environment
• Added Gnome environment selection option (recommended)
• Developer Desktop Package option
• Choosing a keyboard layout
• Ability to enable SSH
• Removed runlevel 5 setting
• Fixed bugs with MOTD
• Turned off automatic update ON
• Improved SSO support
• Optimization script improvements, now called optimize.sh

Tags: VMware, Labs, VDI, Horizon, Linux, Update, VMachines

Recall also that you no longer have the vCenter Server for Windows installer. , vSphere 6.7 - it was latest version platforms where vCenter still had a Windows version. Now it is only a virtual appliance vCenter Server Appliance (vCSA) based on Photon OS.

Earlier we wrote that using the utility that appeared in , you can migrate an external Platform Services Controller (PSC) server to an easy-to-manage embedded PSC using the vCenter Server CLI command interface or the vSphere Client graphical client:

The vCenter 7 installer also upgrades vCenter and transfers all services to the Embedded PSC as part of a single task, so the upgrade will be completed immediately. The new vCenter 7 installer does not have an option to deploy an external PSC:

2. Migration process

If you are migrating from vCenter Server for Windows to vCenter Server Appliance (VCSA), then the scheme will be exactly the same - in the end you will get vCenter 7 on vCSA in an embedded PSC:

Once the external PSC has been converted, it will remain in the console and its decomission is the next task for the vSphere administrator. This can be done using the CMSSO-UTIL command or from GUI client (in the System Configuration section):

3. Ways to upgrade

Everything is simple here. The upgrade is supported according to this plate:

As you can see from the table, the upgrade is supported starting from vSphere 6.5, but many administrators prefer to deploy vCenter services again when upgrading their virtual infrastructure, so as not to drag along the history of possible bugs that may appear during the upgrade.

Before upgrading, you must definitely look at the documents and. But remember that before the official release of vSphere 7, these documents do not contain up-to-date information about the seventh version.

Tags: VMware, vCenter, Upgrade

Now there is an opportunity to override policies. Computer-based policies are applied at system startup. With value RefreshInterval you can control how often these settings are updated before the user logs into the system. And with the value ContinueRefreshAfterLogon you can continue to update the settings after the user login.

Well, the final interesting new feature in DEM 9.11 is Find Items. It will allow you to search in the configuration templates available in the Marketplace, in the Horizon Smart Policies you have created, in a certain set of conditions (condition set) and other elements, which is very convenient for administrators:

You can download Dynamic Environment Manager 9.11 from this link. Release notes are available.

Tags: VMware, DEM, Update, VDI, EUC
Tags: VMware, Horizon, Update, VDI, DEM, Client, EUC

Let's take a look at what's new in vRealize Operations 8.1:

1. Operations with integrated vSphere and Kubernetes infrastructure.

vRealize Operations 8.1 allows you to discover and monitor Kubernetes clusters within a vSphere-integrated infrastructure with the ability to auto-add Supervisor Cluster objects, namespaces (Namespaces), nodes (PODs), and clusters as soon as you add them to vCenter using Workload Management features.

This will give you access to the Summary pages for monitoring performance, capacity, resource usage, and Kubernetes configuration on the vSphere 7.0 platform. For example, Capacity forecasting features will show infrastructure bottlenecks at the node level, while daily operations will benefit from dashboards, reports, views, and alerts.

2. Operations in VMware Cloud on AWS infrastructure.

Now in VMware Cloud on AWS, you can use the VMware Cloud Service Portal token to auto-discover SDDC data centers and set up monitoring tools in a few simple steps. It will also be possible to use one account to manage multiple SDDC objects on the VMware Cloud on AWS platform, including vCenter, vSAN and NSX services, and there will also be full integration with VMConAWS billing.

In the cloud, you can use the following dashboards:

• Monitor resource usage and performance of virtual machines, including NSX Edge, Controller, and vCenter Server services.
• Monitoring of key resources, including CPU, memory, disk and network for the entire infrastructure and virtual machines.
• Monitor resource consumption trends and predict metrics such as Time Remaining, Capacity Remaining, and Virtual Machines Remaining.
• Finding virtual machines that consume unreasonably many resources and require reconfiguration based on historical data.

In addition, VMware NSX-T services will have full support for visualization and monitoring tools:

Well, in the vROPs 8.1 release, there is a full integration of the VMware Cloud on AWS cost tracking functionality with the vRealize Operations solution in the portal interface. This will allow you to control the costs already made and deferred, as well as detail them by subscriptions, consumption and due dates.

The AWS migration assessment engine has also been updated to allow multiple scenario results to be saved for further analysis. These scenarios include various options for Reserved CPU, Reserved Memory, Fault Tolerance, Raid Level, and Discounts.

3. Functions for monitoring multiple clouds (Unified Multicloud monitoring).

Monitoring tools now provide even more advanced features such as Google Cloud Platform support, improved AWS support, and a new Cloud Health Management pack.

vROPS 8.1 now has the following GCP services:

• Compute Engine Instance
• storage bucket
• Cloud VPN
• Big Query
• Kubernetes Engine

The AWS Management Pack now supports the following AWS Objects:

• Elastic Beanstalk
• Direct Connect Gateway
• Target Group
• Transit Gateway
• Internet gateway
• Elastic Network Interface (ENI)
• EKS Cluster

The CloudHealth Management Pack has also been enhanced to include the ability to push GCP prospects and pricing data to vRealize Operations 8.1. You can also create any number of custom dashboards by combining prices for different ratios of public, hybrid or private cloud resources.

As expected, vRealize Operations 8.1 will be released in April this year, simultaneously with the release of VMware vSphere 7. We will definitely write about it.

Tags: VMware, vRealize, Operations, Update, Monitoring, vSphere, Cloud
Tags: VMware, vCenter, VEBA, Labs
Tags: VMware, SRM, Update, DR, Replication, Enterprise

Let's say right away that this is just an announcement, and not an announcement about the availability of a new version of the product for download - as a rule, the GA version of vSphere appears within a month after the announcement. Therefore, we will wait for VMware vSphere 7 in April for the time being, and today we will talk about the new features of this platform.

1. Improvements to VMware vCenter Services

Here you can note the simplification of the vCenter Server SSO topology:

• Ability to upgrade vCenter Server for users with an external PSC to a consolidated topology based on a single vCSA server.
• Embedded PSC is now the only possible deployment option. External PSC is no longer supported.

vCenter Server Profiles:

• This new feature for vCenter servers works exactly the same as Host Profiles works for hosts. You can now compare and export vCenter server settings in JSON format for backup purposes or apply those settings to another vCenter server via the REST API.

vCenter Multi-Homing Features:

• Up to 4 vNICs can now be used for vCSA management traffic, with one vNIC reserved for the vCHA mechanism.

Content Library Improvements

• There is now a new template management view that provides Check-In and Check-Out functions for managing template versions and the ability to roll back to a previous version.
• First, a Check-Out is done to open the possibility of making changes, then a Check-In can be done to save the changes in the library.

New vCenter Feature server update planner:

• The new capability is available as part of vSphere Lifecycle Manager (vLCM) for vCenter Servers.
• With the Upgrade Scheduler, you can be notified about vCenter upgrades, schedule upgrades, roll them out, and perform what-if analysis before upgrading.
• Ability to perform pre-upgrade checks for the selected vCenter server.

2 Improvements to the VMware DRS mechanism

• DRS now runs every minute instead of every 5 minutes as before.
• To generate recommendations, the VM DRS score mechanism (aka ) is used.
• Now this is a Workload centric mechanism - this means that now the needs of the virtual machine itself and the application in it are taken into account first of all, and only then the use of host resources.
• Memory calculations are based on granted memory instead of cluster standard deviation.
• The Scaleable Shares mechanism has appeared, which allows you to better allocate Shares in the resource pool in terms of their balancing.

3. vMotion improvements

Here are the improvements:

• Improvements to Monster VM migrations (with large resources and very high load), which can increase the chance of a successful migration.
• Using only one vCPU when tracking changed pages (page tracer) instead of all vCPUs, which has less impact on performance during migration.
• Reduced context switch time to another server (now less than one second). Achieved by switching when the compacted memory bitmap has already been transferred to the target server, instead of waiting for the full bitmap to be transferred.

4. New Features in vSphere Lifecycle Manager (vLCM)

There are 2 improvements here:

• Cluster Image Management feature, which includes firmware updates, drivers and ESXi images of different versions.
• Initial support for Dell OpenManage and HP OneView solutions.

5. Application Acceleration Features (Tech Preview)

These features come from the acquired company Bitfusion. They allow you to optimize the use of the GPU in the pool over the network, when the vGPU can be partially shared between several VMs. This can be used for AI/ML application task workloads.

All this allows you to organize computing in such a way that ESXi hosts with GPU hardware modules run virtual machines, and their VM companions on regular ESXi servers run applications directly. At the same time, CUDA instructions from client VMs are transmitted to server VMs over the network. You can read more.

6. Assignable Hardware Functions

This feature allows the use of so-called Dynamic DirectPath I/O for machines that need to work with PCIe passthrough and Nvidia GRID devices. It can now be used to match hosts with specific hardware requirements such as vGPU and PCIe. This allows, in turn, to use HA and DRS Initial Placement technologies for such VMs in a cluster where there are ESXi hosts that are compatible in hardware.

7. Certificate Management

There are 2 major new features here:

• New certificate import wizard.
• Certificate API for managing certificates using scripts.

8. Features of Identity Federation

ADFS features are now supported out of the box, and more IDPs using the OAUTH2 and OIDC mechanisms will also be supported.

9. Functions of vSphere Trust Authority (vTA)

• vTA uses a separate cluster of ESXi hosts to create a separate hardware trust node.
• This cluster will be able to encrypt the compute cluster and its VM along with vCenter and other management components.
• You can use the attestation mechanism when encryption keys are required.
• It is now easier to enforce the principle of least privilege and expand the scope of the audit.

10. vSGX/Secures Enclaves capability (Intel)

• Intel Software Guard Extensions (SGX) allow you to move sensitive application logic and storage to a protected area that guest OSes and the ESXi hypervisor cannot access.
• SGX features exclude the use of vMotion, snapshots, Fault Tolerance and other technologies. Therefore, SGX is best used only when there is no other way.

11. New edition of vSphere with Kubernetes (Project Pacific)

We talked about Project Pacific in detail. It is a set of tools for converting a VMware vSphere environment into a native platform for Kubernetes clusters. vCenter Server provides the ability to manage k8s clusters (any clusters older than n-2 will be upgraded). Also integrated into the solution is Harbor, which can be enabled for each namespace.

This is only available for VMware Cloud Foundation (4.0) users so far, as the solution is tied to the .

12. VMware Tools Improvements

Guest Store features are now available in the guest OS (such as updating VMware Tools from the guest OS).

13. Updated hardware (VM Hardware v17)

Here are the main improvements:

• Virtual Watchdog Timer - now there is no dependency on the physical hardware to restart the VM in case the guest OS is not responding.
• Precision Time Protocol (PTP) - for very time sensitive applications (such as trading platforms for traders) you can use PTP instead of NTP and assign its use to virtual machines.

14. vSphere Client improvements

Here are the following improvements:

• The search history began to be saved.
• API Explorer now has a better view of all available APIs.
• Code Capture now has a choice of scripting language - PowerCLI, Javascript, Python or Go.

Of course, this is not all the new features of VMware vSphere 7, presented recently. In the near future we will tell you a lot more about them, and in addition, we will also look at the announced solutions from the VMware Tanzu family, VMware Cloud Foundation 4 and vRealize 8.1.

Tags: VMware, vSphere, Update, Enterprise, Kubernetes, vCenter

To translate virtual addresses into physical ones, a Page Table containing PTE records (Page Table Entries) is used:

PTE records store links to real physical addresses and some parameters of the memory page (you can read more about it). PTE record structures can be of different sizes - they are WORD (16 bits/2 bytes), DWORD (32 bits/4 bytes) and QWORD (64 bits/8 bytes). They address large blocks of addresses in physical memory, for example, a DWORD addresses a block of addresses of 4 kilobytes (for example, addresses from 4096 to 8191).

Memory is read and transferred to the guest system and applications in 4 KB or 2 MB pages - this allows you to read the contents of memory cells in blocks, which significantly speeds up performance. Naturally, with this approach, there is memory fragmentation - rarely when you need to write an integer number of pages, and part of the memory remains unused. As the page size increases, their fragmentation also increases, but speed increases.

Page tables (and there may be more than one) are managed by a software or hardware component of the Memory Management Unit (MMU). In the case of a hardware MMU, the hypervisor transfers broadcast control functions to it, and a software MMU is implemented at the VMM level (Virtual Machine Monitor, part of the ESXi hypervisor):

An important component of the MMU is the Translation Lookaside Buffer (TLB), which is a cache for the MMU. TLB is always located at least in physical memory, and for processors it is often implemented at the level of the CPU itself, so that access to it is as fast as possible. Therefore, the typical TLB access time on the processor is about 10 nanoseconds, while physical memory access is about 100 nanoseconds. VMware vSphere supports Hardware MMU Offload, that is, the transfer of memory management functions to the MMU side of the physical processor.

So, if a request came from the virtual machine to access the virtual address 0x00004105, then this address is split into the virtual page address (Virtual page number - 0x0004) and offset (Offset - 0x105- the area inside the page that is accessed):

The offset is directly passed when accessing a physical page of memory, but the virtual page tag is looked up in the TLB. In this case, there is an entry in the TLB that the address of the physical page corresponding to this tag is 0x0007, respectively, the translation of the virtual page into the physical one was successful. It is called TLB Hit, i.e. cache hit.

Another situation is also possible - when decomposing a virtual address, the resulting tag 0x0003 missing from TLB. In this case, the page is searched for in physical memory by tag (page number 3) and its address is already translated ( 0x006). Next, an entry with this tag is added to the TLB (while old entries are evicted from the cache if it is full):

It should be noted that such an operation causes a slightly higher delay (since it has to be searched in global memory), and this situation is called TLB Miss, that is, a TLB miss.

But this is not the worst situation, since the latency count still goes to nanoseconds. But access can be much longer (milliseconds and even seconds) if the page needed by the guest OS is swapped to disk.

Let's look at an example:

The virtual machine accessed the virtual address 0x00000460, for which there is a tag 0x0000. In physical memory, page 0 is allocated for this tag, which means that you need to look for this page on the disk where the page was flushed due to lack of physical RAM.

In this case, the page is restored from disk to RAM (displacing the oldest page in terms of access time), and then the address to this page is translated. This situation is called page failure ( Page Fault), which leads to delays in application operations, so it is sometimes useful to monitor the Page Faults of individual processes in order to understand the reason for the performance degradation when working with memory.

Tags: VMware, vSphere, ESXi, Memory, Performance, Blogs

Existing vSphere Platinum users after the announced date will receive vSphere Enterprise Plus licenses, VMware AppDefense SaaS product and VMware AppDefense Plugin for vSphere (where to download this plugin is written). For vCloud Suite Platinum and Cloud Foundation Platinum users, nothing changes, except for the evolution of vSphere itself, which is part of the packages.

Tags: VMware, vSphere, Platinum, Update, Support

The package focuses on code quality, code reuse, unit testing, relationship management, and parallel project releases for the vRealize platform. vRealize Build Tools are extensions packaged in the Maven repository format that support the use of an IDE (via Maven) as well as a CLI interface for developing, testing, and deploying solutions for vRA/vRO platforms.

Let's see what's new in the second version:

• Support for the solution, its blueprints, custom forms, subscriptions, and flavor-mapping mechanics
• Support for existing content and import it for vRO 8
• Support for vRO 8 features to export workflows to a folder structure based on their tags
• Running workflows on vRO using maven command
• Ability to save JS Actions IDs on origin to prevent conflicts in vRO environment
• Improvements to experimental support for TypeScript projects
• Bug fixes and documentation updates

To get started with vRealize Build Tools, you will need the following tools:

• vRealize Orchestrator
• Microsoft VS Code

You can download vRealize Build Tools from this link.

Tags: VMware, Labs, vRealize, Automation, Orchestrator, Update

In addition to many bug fixes, the utility has several new cmdlets:

• Add-vRA-Project-Administrator
• Add-vRA-Project-Member
• Get-vRA-DeploymentFilters
• Get-vRA-DeploymentFilterTypes
• Get-vRA-FabricNetworksFilter
• Get-vRA-FabricImagesFilter
• Remove-vRA-Project-Administrator
• Remove-vRA-Project-Member
• Update-vRA-Project-ZoneConfig

Recall that this module is not supported by VMware (like all utilities on VMware Labs that are in the Tech Preview status), so use it carefully.

This tool may be useful to you in the following cases:

• When you need to compare two clusters in terms of performance (for example, on different hardware)
• When to Understand the Impact of Cluster Configuration Changes on Performance
• When to check if a new cluster is configured correctly before launching it into a production environment

To run Weathervane, you need to create container images, prepare a configuration file, and run a benchmark. Further, the utility itself will deploy containers in the cluster, launch applications and collect test results.

Weathervane deploys the benchmark application on the nodes and feeds the load there, which is generated through the Workload driver component. This driver can be located either together with the benchmark application or in an external environment, in a separate cluster.

Weathervane can be set to run at a constant load for a fixed number of simulated users, or it can be configured to look for the maximum number of users so that quality-of-service (QoS) requirements are met. In the latter case, the result of the test will be the maximum number of WvUsers that the cluster can support. Actually, this parameter should be used to compare clusters in terms of performance.

Here are the components of the Weathervane solution (the Run harness component is responsible for executing the test runs and getting the test results):

Weathervane uses a multi-tiered web application that includes both stateless and stateful services. You can choose from one of these application deployment types. Multiple application instances can be run in a single run, allowing testing to scale across large clusters.

The Weathervane app has several tiers. The application logic is implemented through Java services running on the Tomcat server, which communicate via the REST API and RabbitMQ messages, and Zookeeper is used for coordination. Backend storages are implemented using PostgreSQL and Cassandra. Frontend web servers and proxy cache servers are implemented on Nginx.

Tags: VMware, Kubernetes, Weathvane, Update, Performance

In Russia, there are already 10 vExpert carriers, not so many, but not enough (at the level of Sweden and Norway). It is clear that the majority of vExpert is from those countries where everything is fine with English, since the audience of blogs on English language more broadly, which motivates authors to write posts (and in general, vExpert is given for blogging).

This is what the top ten looks like:

And here are those specialists from Russia who received vExpert this year:

Tags: VMware, vExpert, Blogs

VMware vCenter Server 6.7 server performance when working with a virtual infrastructure of VMware ESXi servers of remote offices and branches

Many users of the VMware vSphere platform know that there is such an option for deploying and operating a distributed virtual infrastructure as ROBO (Remote or Brunch Offices). It implies the presence of one or more main data centers, from where small remote offices are managed, hosting several VMware ESXi servers with or without their own vCenter.

At the end of last year, VMware released an interesting document "Performance of VMware vCenter Server 6.7 in Remote Offices and Branch Offices" ( we have already talked about it a little), which discusses the main aspect of using such a scenario - performance. After all, remote offices can be located in other cities, countries and even continents, which are accessed via different types of connections (for example, 4G or satellite), so it is very important how much traffic various operations consume, and how quickly they work out from the point of view of the administrator.

Parameters of various types network connections in VMware they put it into a table (in the right column, which was the result of using a test configuration, and in the left column, as happens in scenarios with real data centers):

For testing, a remote configuration of 128 ESXi hosts was used, where 3840 virtual machines were registered (960 VMs per cluster, 30 per host), of which up to 3000 machines were turned on simultaneously.

Hypervisors (virtualization technologies) have existed for more than 30 years and during this time have managed to become one of the main "cogs" in the cloud ecosystem. Many companies looking for virtualization solutions choose two popular hypervisors - VMware and KVM. Let's find out which one is better. But first, a little theory.

What is a hypervisor?

A hypervisor is a program that separates the operating system from the hardware. Hypervisors virtualize server resources (processor, memory, disk, network interfaces, etc.), allowing them to be used as their own, and create several separate virtual machines based on one server. Each created virtual machine is isolated from its neighbors so as not to affect the work of others. For the hypervisor to work, virtualization support is required: for Intel processors on an Intel VT processor, and for AMD processors on AMD-V.

Hypervisors are divided into two types: the first work directly with the server, and the user's operating system runs on top of the hypervisor. These hypervisors can provide server management functionality to some users, and most enterprises use these hypervisors.

Hypervisors of the second type, also known as hosted hypervisors (Hosted Hypervisor), work with the operating system installed on the server. And operating systems for new users are built on top of the hypervisor.

Desktop hypervisors such as Oracle VirtualBox or VMware Workstation are type 2 hypervisors, while VMware and KVM are type 1. VMware and KVM are installed directly on the server and do not require any operating system to be installed.

VMware vSphere

Before buying VMware vSphere, you can try to work in a trial version (60 days), after which you need to buy a license, or put up with the limitations of the free version.

The free version, called VMware Free vSphere Hypervisor, has no CPU and memory limits per host, but a number of others:

• The product API is read-only;
• a virtual machine cannot have more than 8 cores;
• it cannot be used with Veeam to create backups;
• connection to vCenter Server is not supported;
• high availability is not supported, as are VM Host Live Migration and VM Storage Live Migration technologies.

The product from VMware differs from analogues by supporting a large number of operating systems - Windows, Linux, Solaris, FreeBSD, Netware, MacOS and others.

Installing a VMware distribution on a server is very simple: just boot from a CD, flash drive, or via PXE. In addition, scripts are supported to automate the process of installing software, configuring the network, and connecting to the vCenter Server.

It is also important to have a special converter VMware vCenter Converter, which allows you to use MS Virtual Server, Virtual PC, Hyper-V images in ESXi, as well as physical servers and disk partition images created by programs such as Acronis True Image, Norton Ghost and others.

VMware vSphere has built-in integration with Microsoft Active Directory, which means that users in a private or hybrid cloud can be authenticated using Microsoft Domain Services. Flexible resource allocation allows for hot-adding of CPU, RAM and hard drive(including resizing the current hard drive without rebooting).

VMware Fault Tolerate is a VMware technology designed to protect virtual machines using continuous availability clusters. If the host (ESXi server) with the primary (Primary) working copy of the virtual machine fails, the protected virtual machine will instantly switch to the "secondary" (Secondary) or "shadow" copy running on another ESXi server. For machines protected by VMware Fault Tolerance, there is a constant (real-time) copying of the entire state of memory and processor instructions from the main copy to the "shadow". If the primary ESXi host fails, users will not even notice the failover process to the second host. This is where Fault Tolerance differs from High Availability. In High Availability, if a physical server fails, the virtual machines will be restarted on other nodes, and while the operating systems are rebooting, users will not be able to access the virtual servers.

In addition to VMware Foult Tolerate, the VMware vCloud Suite Enterprise license provides high availability, fault tolerance, and disaster recovery with vSphere HA, vMotion, Storage vMotion, and vCenter Site Recovery Manager features.

To reduce planned interruptions in the maintenance of servers or storage systems (SAN), the vMotion and Storage vMotion functions online migrate virtual machines and their disks without stopping applications and users. The vSphere Replication feature supports various vCenter Site Recovery Manager (SRM) replication options to protect against major disasters. SRM provides centralized disaster recovery planning, automatic Failover and Failback from a failover site or vCloud, and non-disruptive disaster recovery testing.

The features of this hypervisor include selectivity to hardware - before installation, you must carefully check the existing equipment for compatibility with the desired version of ESXi. For this, the VMware website has a special .

Licensing of VMware products has its own peculiarities. Additional confusion is added by periodic changes (from version to version of vSphere) in the licensing policy of VMware. There are a few things to consider before purchasing VMware vSpere licenses:

• Hypervisor licensing is based on the number of physical processors (CPUs). Each server CPU requires a separate vSphere license (cores are not physical processors and do not count towards licensing);
• the available functionality of the ESXi server is determined by the vSphere license installed on it. Detailed guide according to licenses there are on;
• for each vShpere license purchased, you must purchase a service support package (at least for a year);
• VMware does not impose limits on the amount of memory (RAM) installed on a server and on the number of running virtual machines.

You can manage multiple hosts with ESXi hypervisors, storage systems and network equipment using another VMware product - Vcenter Server. vSphere client plugins provided by VMware partners give IT administrators the ability to manage third-party elements in the data center directly from this console. Therefore, vCenter users can backup, protect data, manage servers, networks, and security directly from the vCenter interface. In the same console, you can set up triggers that will notify you of problems that have arisen, and get data on the operation of the entire infrastructure in the form of graphs or tables.

KVM

KVM is an easy-to-use, lightweight, low-resource, and fairly functional hypervisor. It allows you to deploy a virtualization platform in the shortest possible time and organize virtualization under the control of the Linux operating system. During operation, KMV accesses the operating system kernel through a special module (KVM-Intel or KVM-AMD). Initially, KVM only supported x86 processors, but modern versions KVM is supported by a wide variety of processors and guest operating systems, including Linux, BSD, Solaris, Windows, etc. By the way, all Wiki resources (MediaWiki, Wikimedia Foundation, Wikipedia, Wikivoyage, Wikidata, Wikiversity) use this particular hypervisor.

Because the guest operating systems interact with the hypervisor that is integrated into Linux kernel, guest operating systems have the ability to access hardware directly without having to modify the guest operating system. Due to this, there is almost no slowdown in the guest operating system.

KVM allows virtual machines to use unmodified QEMU, VMware and other images containing operating systems. Each virtual machine has its own virtual hardware: network cards, disk, video card and other hardware.

Thanks to the support of unmodified VMware images, a physical server can be easily virtualized using the same VMware vServer Converter utility, and then transferred to the hypervisor.

Installing KVM in the operating room Linux system consists in installing the KVM package and the Libvirt virtualization library, as well as carefully setting up the virtualization environment. Depending on the operating system used on the host, you need to set up a bridge or connection to a VNC console through which the virtual machines will communicate with the host.

Administering KVM is more difficult, as transparent access to files, processes, consoles, and network interfaces missing, you have to configure it yourself. Rebuilding VM parameters in KVM (CPU, RAM, HDD) is not very convenient and requires additional steps, including rebooting the OS.

The project itself does not offer convenient graphical tools for managing virtual machines, only the Virsh utility that implements all the necessary functions. For convenient management of virtual machines, you can additionally install the Virt-Manager package.

KVM does not have built-in tools like Fault Tolerate for VMware, so the only way to create a high availability cluster is to use network replication using DRDB. The DRBD cluster only supports two nodes, and the nodes are synchronized without encryption. That is, for a more secure connection, you must use a VPN connection.

In addition, to build a high-availability cluster, you will need the Heartbeat program, which allows you to exchange service messages about your state between nodes in the cluster, and Pacemaker, a cluster resource manager.

The KVM hypervisor is distributed as an open source product, and for corporate users there is a commercial Red Hat Virtualization (RHEL) solution based on KVM and the oVirt virtual infrastructure management platform.

The undoubted advantage of this hypervisor is that it can run on any server. The hypervisor is quite unpretentious in terms of resources, which makes it easy to use it for testing tasks.

Please note that KVM does not have a helpdesk. If something does not work out, you can count on the forums and the help of the community. Or switch to RHEL.

So what to choose?

Both hypervisors are mature, reliable, high-performance virtualization systems, each with its own characteristics to consider when choosing.

KVM is usually more scalable than VMware, primarily because vSphere has some limitations on the servers it can manage. In addition, VMware has added a large number of Storage Area Networks (SANs) to support multiple vendors. This feature means VMware has more storage options than KVM, but it also makes it harder to support VMware storage when expanding.

KVM is usually the most popular hypervisor for companies looking to reduce implementation costs and less interested in enterprise-grade features.

Studies have shown that KVM's total cost of ownership is typically 39 percent lower than that of VMware, although the actual total cost of ownership depends on specific factors such as operational parameters and site workload.

Tight integration with the operating system on the host is one of the most common reasons developers choose KVM. Especially those who use Linux. The inclusion of KVM in many Linux distributions also makes it a convenient choice for developers.

Cloud providers offering IaaS services to their customers usually choose an infrastructure built on VMware products. Solutions based on VMware Sphere contain all the important enterprise features for high and continuous availability, provide support for more guest operating systems and have the ability to interface the customer's infrastructure with cloud services.

vSAN 6.7 improves HCI operational efficiency, reduces training time, and accelerates decision making. This release provides more consistent, resilient, and secure application support. In addition, the knowledge of leading experts, the latest technologies and analysis tools are used to resolve problems more conveniently and faster. More companies and cloud service providers are choosing VMware vSAN as their hyperconverged infrastructure solution.

OPPORTUNITIES

Product Improvements

Below are the major new features and updates in vSAN 6.7.

• HTML5 based user interface
• Completely updated user interface provides modern control options. The new interface was built on the same platform as used in other VMware products, giving customers a unified and seamless experience for managing the most complete SDDC product stack. In addition, the new interface reduces the number of steps required to complete many tasks by streamlining workflows.
• vRealize Operations in vCenter
• Integrated directly into vCenter, vRealize Operations provides complete visualization of HCI environments deployed on-premises or across a wide range of public clouds and is available free of charge to all vSAN Advanced and Enterprise edition customers. With customizable vSAN dashboards in a single management console, you can monitor and control your HCI environment. Integrating new or existing vROPs is non-disruptive.
• vSAN ReadyCare
• vSAN ReadyCare Support underscores VMware's commitment to vSAN customers and provides end-to-end support with leading experts and the latest analytics and technologies. Using predictive modeling in vSAN Support Insight, VMware analyzes anonymously collected data from thousands of vSAN customers and sends notifications to them before problems occur. In addition, vSAN health check services provide real-time notifications and troubleshooting recommendations.
• FIPS 140-2 Encryption
• vSAN implements the first industry-standard data-at-rest encryption solution for HCI. vSAN 6.7 introduces vSAN Encryption, the first software solution to comply with the FIPS 140-2 standard and stringent US federal government requirements. vSAN Encryption reduces data protection costs and increases agility by being hardware-free and simplifying key management processes. It is also the first HCI solution to have STIG approved by the Defense Information Systems Agency (DISA).
• Enhanced application resiliency
• The vSAN solution provides a consistent end-user experience with applications through intelligent self-healing capabilities such as adaptive resynchronization, fast failover for physically separated networks, and replica consolidation. Adaptive Sync optimizes I/O traffic management to keep applications running while syncing. Consolidating replicas reduces the time and effort required to bring a node into maintenance mode. Finally, the need for failover of physically separated networks is eliminated with instantaneous failover.
• Increased availability of distributed clusters
• Distributed cluster environments are made even more efficient by intelligently separating witness component traffic, redefining the primary environment, and efficient resynchronization. Witness traffic separation and efficient synchronization optimize the route and size of the data that is sent on each link, making failover transparent to application end users. Redefining the primary environment improves the availability of workloads by using more efficient logic in the event of an environment failure.
• Optimized Next Generation Apps
• The vSAN solution uses a new storage policy (vSAN host pinning) to bring vSAN efficiency and resiliency in line with the needs of the latest shared-nothing applications. By applying this policy, vSAN keeps one copy of the data and writes blocks of data on the ESXi host that is running the VM. This feature is especially important for applications that work with large amounts of data (Hadoop), NoSQL (DataStax), and other applications that perform application-level data backup.
• Extended Support for Critical Business Application Environments
• vSAN now supports more mission-critical application environments with support for Windows Server Failover Clustering, simplifying storage management for these workloads and helping customers accelerate their transition to a centralized SDDC.
• Proactive support with vSAN Support Insight
• Proactive support improves vSAN reliability with alerts that are generated before infrastructure issues occur, and reduces traditional support time with periodic data collection. To use this feature, you must enroll in the User Experience Improvement Program.
• Support for Adaptive Core Dump
• Adaptive Core Dump support reduces the resolution time for vSAN customers for more types of environments by automatically adjusting the direction and size of valuable data used to speed up support.
• Extended hardware support
• vSAN now supports 4Kn drives to help prepare vSAN environments for future needs and provide an opportunity to reduce total cost of ownership.