Virtual machines, containers, functions. Market knowledge for IT decision makers since 2003
As expected, VMware has launched a new classroom course about the automation of the vSphere virtual infrastructure through the Microsoft PowerShell implementation called PowerCLI.
The course objectives are:
In mid March Microsoft finally announced its plan to integrate the Calista technology, acquired in January 2008, in the upcoming Service Pack 1 for Windows Server 2008 R2.
The technology, dubbed RemoteFX, extends the Remote Desktop capabilities and leverages the presence of supported GPUs on the physical servers that are used for VDI.
So far the company didn’t share much details about all the capabilities of RemoteFX 1.0, but a couple of weeks ago, during its TechEd 2010 conference, the company unveiled at least one thing: the support for generic USB redirection.
Bink.nu reports about the details:
Now that it’s near the release of its vCloud Service Director (vCSD, formerly project Redwood) VMware is working to focus the attention on its programming interfaces and ignite an ecosystem of new developers that can build on top of the vCloud platform.
Part of this effort consists in the launch of a new SDK support program.
Where VMware can help?
To further facilitate the evaluation of its new client hypervisor, Citrix has recently published a Proof of Concept Implementation Guide which guides customers step-by-step in the setup of a VDI infrastructure that include XenClient, Receiver and Synchronizer.
The 53-pages documentation includes guidance to test the experimental feature called Secure Application Sharing, already available as part of the XenClient Express 1.0 Release Candidate, which is used to share data between virtual machines.
The upcoming release 4.1 of vSphere will introduce a new memory over-commitment technique called Memory Compression which VMware claims to greatly improve virtual machines performance under heavy load.
There may be further innovation in this area coming in the near future: in January 2010 a number of Chinese researchers presented a new technique called Dynamic Memory Paravirtualization (DMP) for virtual infrastructures:
In dynamic paravirtualization, VMM (virtual machine monitor) dynamically monitors and replaces the hot instructions, which cause most VM exits. It is transparent to the guest OS such that the legacy OSes can benefit from this optimization. Our study focuses on reducing the overhead of memory virtualization—dynamic memory paravirtualization (DMP). We implant a new memory management mechanism in VMM such that all user-mode page faults can be handled by the guest OS directly without VM exits. We implement a prototype of dynamic memory paravirtulization based on a version of KVM using Intel VT. Our experimental results show that our technique essentially eliminates the overhead of VM exits caused by page faults. Dynamic memory paravirtualization can achieve the effectiveness of paravirtualization without changing the source code of guest OS.
A few days ago IBM published a very interesting study about the support model used by Amazon for its Infrastructure-as-a-Service cloud platform: EC2.
The 7-pages research summarizes the analysis of the Amazon support forums between August 2006 and December 2009, and highlights how the cloud provider is having issues in dealing with a lot of troubleshooting:
…We found that, with the exception of problems related to application-level issues, the observed problems are roughly evenly divided among the remaining problem classes (e.g., connectivity, virtual image management, performance, etc.). In studying the evolution of problems over time, we find that some classes of problems are closely related to the introduction of new cloud features as users incorporate them into their deployments. Other problems, such as those related to connectivity, are persistent and are relatively less affected by new functionality. We see evidence that some problem classes (e.g., related to image management) diminish significantly as cloud providers introduce new interfaces and tooling to simplify certain operations. The level of involvement from the cloud operator in solving problems also is dependent on the problem class, and changes over time. Some classes, such as those problems related to virtual infrastructure components, require operator involvement 50% of the time. While operator involvement in solving problems does generally decrease over time, some problems consistently require help from the cloud operator…
Like any other virtualization vendor, of course Microsoft is actively researching to develop more efficient and secure architectures for its hypervisor.
While most of this work remains undisclosed in the Research labs, a few things are being shown in public events.
It’s the case of Bunker-V, the codename for a new architectural approach that Microsoft is evaluating to reduce the Hyper-V trusted computing base (TCB), which today includes the virtual machine monitor (VMM) and the parent partition, equal to over 50 million of lines of code.
The research appeared in a speech titled Improving the Security of Commodity Hypervisors for Cloud Computing presented this year at the Seventh Annual Microsoft Research Networking Summit.
The slide deck describes Microsoft is redesigning its hypervisor to be faster and with a smaller TBC for cloud computing scenarios.
As a few readers may know, Google is fervently working to launch its own operating system. Called Chrome OS and released as an open source project, it uses the Chrome browser as its core engine.
Announced in July 2009, Chrome OS is expected to reach version 1.0 within the 2010 holiday season, and to be deployable on x86 and ARM architectures (including upcoming tablet PC devices that will compete with the recently released Apple iPad).
The brave ones that want to test the alpha versions of the OS can do that using most hardware virtualization platforms on the market. Parallels even offer support for it.
The more we get near the planned release timeframe the more details are shared about how it will work.
Fresh information appeared a few days ago about a new feature tentatively dubbed Chromoting.
At virtualization.info we usually don’t cover when a hosting provider adopts one virtualization or cloud platform rather than another, but of course there are a few exceptions.
Today we have one, with the announcement of Softlayer offering Parallels Server 4.5 Bare Metal as part of its virtualization portfolio.
The hosting provider already offers Virtuozzo Containers, the Parallels OS virtualization technology, since some time.
Parallels launched its hypervisor in October 2009 after several delays. So far the company didn’t spend much effort in promoting it and just a few customers are really aware that it’s an alternative to Citrix XenServer, Microsoft Hyper-V, Oracle VM Server and VMware ESX/ESXi.
The fact that Parallels Server Bare Metal is the only product in its category that doesn’t offer a free version doesn’t help much.
In November 2009 VMware released version 1.1 of its disk & file backup/restore solution for guest operating system dubbed Data Recovery. The company released last week a new minor update.
Data Recovery 1.2 (build 260251) introduces a major new feature: the File Level Restore (FLR) support for Linux guest OSes.
The new build also introduces a few enhancements to the vSphere Client plug-in but the most important one probably is that now each vCenter Server can support up to 10 Data Recovery virtual appliances.
And of course the build solves a long list of bugs.
While the product is slowly improving from a technology standpoint, VMware continues to avoid promoting it, maybe because it’s not mature enough to compete with the well established solutions from PHD Virtual, Veeam and Vizioncore, or maybe because it doesn’t want to disturb too much these partners and push them into the arms of Microsoft and Citrix.