Just what are the important performance factors for Virtualization?
November 21, 2007 – 4:20 pmSteve Wilson does a good job at putting the Oracle benchmarketing claims into persepective. Later in his post, he touches on a few factors which are more relevant in the practical world, including power-performance ratios.
There are many aspects by which we can measure virtual-performance. I strongly agree that alternate factors like power are becoming very important, and that there are several more that are typically considered. Historically, the primary factors have been:
- CPU Efficiency: How much CPU is used to deliver a prescribed throughput
- Price-performance: Performance vs. cost of the CPU
Going forward, I think the following will be more important:
- Throughput: can the application deliver the required levels of throughput, in terms of real world transactions?
- Latency: is the latency of each transaction within tolerances, or affected by virtualization
- Scalability: does throughput/latency change as load is increased (often asked in the context of – “do I have enough future headroom?”)
- Memory efficiency: Doing more work with less memory – multi-core is rapidly providing an increasing amount of CPU, but memory remains a premium
- Power-performance: Throughput relative to power, i.e. how much performance can I deliver for the power consumed
- Space-performance: How much performance can I get per rack unit
- Agility: What is the time taken to deploy a new application
Feel free to comment on what the most important factors for virtualization performance are in your environments today…
3 Responses to “Just what are the important performance factors for Virtualization?”
I think you need to add some more “illities”. Observability is one of the biggest problems with virtualization, systems become unpredictable and unmanageable, and have performance related outages just because the administrators can’t figure out how much headroom there is and how it is varying.
Availability is the other one, how much downtime is needed to change various aspects of a VM/OS combination? e.g. Some OS’s need a reboot if the VM adds more memory, others are dynamic. How about upgrades for the host system and VM itself? If you take out ten client VMs every time the host needs a reboot, you have created an issue. There is a trade-off between performance and availability. The more insulated the client VM is, the more available it can be, but there will be higher overhead, and slow periods during online maintenance.
Cheers and happy new year
Adrian Cockcroft
By Adrian on Dec 31, 2007
Adrian I think you’ve answered your own question there. Managers can’t figure it out… If you don’t understand how to configure the host suitable to the guest OS’s then you’ll run into trouble… but then this also is a problem with physical boxes if they’ve been speced wrong, so no difference there.
VM’s need reboot for more memory, so do physical (yes some support it, but it’s rare in the big picture of things)
Host needing reboot and taking out 10 machines, well with VMotion and a SAN you can move them off to another hosts, update and reboot the guest, move them back and you’re off again.
One thing to remember that there will be new skills needed in managing a virtual environment, but they’re not rocket science, it’s more an extension of you’re current skills looking after physical boxes.
I’ve been in this job for over 10 years now and I can say Virtulization is a godsend to the med size businesses. Failover and HA for the masses!
By Raymond Brighenti on May 20, 2008
Hi people!
)
The interesting name of a site – blog.richardmcdougall.com
I today 7 hours
looked in the Internet So I have found your site
The interesting site but does not suffice several sections!
However this section is very necessary!
I have added you in elected!
Forgive I is drunk
By fleeludge on Oct 19, 2008