latoga labs

Tonight VMware has done their online announcement of vSphere 4.  While some may consider the branding of vSphere 4 as a Cloud Operating System just trendy technology marketing, there are enough new features, enhancements to existing capabilities, API hooks to enable new product add-ons, and partnership activity to have justified the re-branding away from technology to provide enterprises with Virtual Infrastructure to a solution that enables enterprises to operate their own cloud computing enrionment…a Cloud OS.

In case you haven’t read the disclosure like over on the right, I am currently employed by VMware as a Global Solutions Consultant.  This means that I have been talking to global customers about vSphere 4 for quite a while now.  And when it takes two hours to give a product road map that just touches on all the capabilities that a release includes, it’s big enough to be worth the re-branding.

Here is my launch day link roundup to VMware vSphere 4 for Enterprises:

• First Off, the Simulcast of the official launch by VMware Executives and Partner Community (9am PDT on April 21st) [Look for me snapping photos during the event, probably from behind a big white lens...stay tuend for more on that tomorrow]
• vSphere list of Key Features (PDF)
• Summary of VMware product name changes
• vSphere is organized into two main sets of services
  • Application Services
    • Availability
    • Security
    • Scalability
  • Infrastructure Services
    • vCompute (ESX and ESXi are still there as the rock solid foundation)
    • vStorage: What’s New Whitepaper (PDF)
    • vNetwork: What’s New Whitepaper (PDF)
• vSphere editions for Enterprises
• vSphere Pricing, Packaging, and Licensing Whitepaper (PDF)
• vSphere Upgrade Center is now also live…start planning your migration from VI3 to vS4!

For non official teaser screen shots of various parts of vSphere 4, check out Jason Boche’s A random collection of what’s new vSphere eye candy.  I’m sure that there will be dozens of other blog posts in the near future talking about the various new or enhanced aspects of vSphere in wonderful technical glory.  I’ll try to share the best ones that I come across and that my customers post via my Twitter feed.

While vSphere contains enough things to keep every member of an IT staff talking for days, my focus over the next few days (besides discussing various parts of vSphere to my clients) will be to talk about the key aspects of vSphere that I think can help an enterprise continue to drive cost savings while enabling agile service delivery to the business, items like:

• Distributed Resource Scheduling and Distributed Power Management as they relate to an internal cloud
• Distributed Virtual Switching and it’s hooks into Application Services like vShield Zones and VMsafe
• Storage savings features of Thin Provisioning
• Fault Tolerance and the hidden revolution this provides
• Easing the Management burden using vCenter Orchestrator and vCenter Server Linked Mode and new licensing model

Enjoy the launch day festivities…

A customer of mine today asked about the the results from a recently run benchmark of hypervisors published by Virtualization Review in which ESX, Hyper-V, and XenServer were compared.  There is a post on the VMware blog questioning the configuration of the benchmark environment, and thus the results and conclusions from the benchmark.  I wanted to share both of these links for those who may hve seen only the report and were scratching their heads as well.

~ ~ ~ ~ ~

Since the benchmark was highly based on SQL server running in a VM, this seems a good time to also share additional research recently done on SQL Server and shared at VMworld 2009 in France.  This performance research sheds some light on that fact that SQL Server Performance Problems are Not Due to VMware.  After hearing lots of customer complaints about poor SQL server performance last year at VMworld in Las Vegas, VMware’s performance team spent three months looking at every way increased performance could be sqeezed out of SQL Server by tweaking ESX, the guest OS, and SQL Server.  The net result was that most performance issues seen in running SQL Server virtualized on ESX come from mis-configurations in the other components and not from ESX.

A few weeks ago VMware internally released a best practices guide for running Java in Virtual Machines on ESX (pdf download).  Recently it was posted on the VMware Support Technical Papers portal.  This document covers subjects like:

• Multi-tiered Java Applications
• Java Heap and Virtual Machine Memory
• Virtual CPUs and Threads
• Disk I/O
• Timekeeping and
• Micro-Benchmarks

Also check out the VROOM! post discuss how ESX Runs Java Virtual Machines with Near-Native Performance.

About a month ago I had a friend contact me asking about some issues he had while buidling a low cost ESX server for his home lab.  He was having difficulty with regards to one small piece of hardware, and he didn’t know about the hardware compatibility list for ESX.  While it’s a bit late for my friend Dave, here is some great instructions on how to build a $500 ESX Server.  Also check out the VM Help site for an unofficial whitebox HCL list; this is not hardware offially supported by VMware but apparently tested by the internet VMware community…your milage may vary.

Last week, before the end of VMworld, we had a session with one of my customer’s to discuss ESX performance.  This discussion was lead by one of VMware’s performance gurus Scott Drummonds.  Scott works as the Manager of VMware’s Performance Marketing team, working with the VMware field teams and customers to provide and advise on product performance issues.

During this conversation, I wrote down some key quotes from Scott that I wanted to share.  For those that have been around VMTN for a while, you might already know some or all of these performance related suggestions.  However, they are key enough that I wanted to draw attention to them.

• “Administration work on the LUN has an impact on performance more than the number of hosts.”

When admin work is done on the LUN is the only time that SCSI reservation locking is used.  So placing more VMDKs on the same LUN doesn’t directly impact the performance.  See Scalable Storage Performance with VMware ESX Server 3.5 Vroom! blog post for additional details.

• “Put Windows VMs and Linux VMs in sperate clusters because they can share memory more efficiently…”

Part of the memory optimization doen within the ESX hyperadvisor is to share common memory used by multiple VMs.  So if you are running 10 Windows VM, the memory used to store the majority of the Windows OS is shared amonst all 10 VMs since the memory contents are the same and do not change.  This is what enables memory overcommitting.  If you start to mix and match different OS VMs on the same host, this advantage can be minimized.

• “High storage latencies is the largest source of performance problems that I see…”

You can monitor the storage latencies within VC by changing the stats level.  See Scott’s Understanding VirtualCenter Performance Statistics performance community doc for more details.

• “Once your swapping, you’re in trouble…”

Seeing swapping within your VM means that you are not allocating enough memory to the VM for the applications that are running.  Watch this closely and add memory or migrate VMs when and where needed.

• “RHEL5 does 1000 interrupts per second for “greater precision” (versus most OS’ which only do 100) which can add up to un-wanted overhead in the OS…”

This is an issue with the Linux Timer Rates that Scott talks about.  There is a small configuration change that you can make in RHEL5 SMP that will provide an across the board increase in system performance.

Check out all of Scott’s VMTN articles, there is a wealth of information in them.