Yellow Bricks

by Duncan Epping

Archives for February 2009

RE: max num vCPU’s Malaysia VMware Communities

· ·

I was just reading this article on the Malaysia VMware Communities website I’ve read a couple of article on their website that didn’t make sense but this this time I’m going to respond cause it might set people on the wrong foot. Anyone is of course entitled to their own opinion and views, but please reread your article and check the facts before you publish, especially when your blog is featured on planetv12n. A short outtake of the blog:

If we refer to the current version which is ESX 3.5 u3, the maximum number of Vcpu per ESX server is only 128 per ESX Servers. Personally, I think the number of Vcpu per ESX servers is too minimal. Imagine if we do run a servers with 4 or 8 physical CPU sockets and we consolidate 40 : 1 Physical server in our virtualization environment, we will hit to the bottleneck on maximum numbers of Vcpu per ESX servers but not due to the CPU consumption

Reading this short section one might think, why reply it makes sense? No, it doesn’t make sense at all:

  • The current limit isn’t 128, it’s 192 vCPU’s.
    So even with a 40:1 ratio and all VMs provisioned with 4 vCPU’s you wouldn’t hit this limit. Read the max config guide, it’s the bible for virtualization consultants.
  • But even more important: co-scheduling and over provisioning will impact performance. With most VM’s running 2 or even 4 vCPU’s scheduling will be almost impossible even with the relaxed co-scheduling techniques ESX is using these days. In other words, please don’t use multi vCPU VMs as a standard, you can read more on c0-scheduling here.

The author asked VMware to bump up the max number of vCPU’s. Now for a VDI environment this can and will be useful I think. Again if you are hitting the number with a 16 core machine, you might need to reconsider your provisioning strategy.

I expect the number to go up… especially after watching Stephen Herrod’s keynote at VMworld Europe 2009.

VMworld Swag…

· ·

One of my former colleagues held a “Swag competition” with a friend from South Africa….
South Africa 0, Netherlands 1.

(ps: this picture doesn’t include Thursdays harvest.)

SRM Patch released

· ·

VMware just released patch 2 for SRM 1.0 Update 1. It’s a cumulative patch that corrects several problems with 1.o Update 1:

  • a problem that prevents protected virtual machines from following recommended Distributed Resource Scheduler (DRS) settings when recovering to more than one DRS cluster.
  • a problem observed at sites that support more than seven ESX hosts. If you refresh inventory mappings when connected to such a site, the display becomes unresponsive for up to ten minutes. (This problem was also addressed in Patch 1 for SRM 1.0 Update 1.)

Before you apply the patch please backup both SRM server’s / databases.

VMworld day 2 – Random stuff and Data recovery

· ·

What a day again, and it’s actually not finished cause within a couple of minutes the VMworld Europe 2009 party will start.

The VMTN Experts session was actually, in my opinion, pretty good today. We had way more people coming over that were asking questions or just came over to have a chat with one of the Experts! With the vExperts being announced this morning there was a good atmosphere, even Statler and Waldorf euuuh Boche and Laverick were having fun. For those on twitter and those who have been following the story about VMDougs snuggie, check the video Gabe published, lol.

Now today I did one of the labs together with Eric Sloof, Data Recovery. There are a couple of write-ups on the product for instance this one by VM-Aware so I’m going to write about all the features and scheduling possibilities. One thing I’ve noticed no one writing about is that the Appliance actually uses the hot-add feature which is part of VCB. So basically what happens when you do a backup:

  1. Create a snapshot of disk(s)
  2. Hot add disk(s) to Data Recovery appliance
  3. Create hashes of (hopefully variable) blocks
  4. Read data of changed blocks
  5. Dedupe(using variable chunk sizes) and create hash
  6. Store data
  7. Remove hot add disk(s)
  8. Remove snapshot

    So this is what I’m guessing will happen cause I actually haven’t seen the documentation, but it makes sense in my opinion.. The cool thing about this way of running backups is that doesn’t really matter if you do a full backup or do an incremental/differential it’s always small because of the deduplication. Full backup it is! Although some blogs aren’t sure yet, it will contain file level restore in the future!

    By using the hot add mode it’s a LAN free backup, all I/O is being handled by the ESX Server I/O stack. This will however cause overhead on the ESX Server so you might not want to start 20 backups on the same host. Also keep in mind that the deduplication will be rather CPU intensive so this might also slow down the process if you run multiple backups. The deduplication is “inline” by the way which means that the data will be deduped before it will be written to disk.

    Now there’s more to write about, and especially about the Cloud Plugin that has been demoed today… But I really need to get ready for the VMworld party called “Cloud9” this year! See you guys in a few minutes 🙂 and for those that didn’t come to Cannes, please join John Troyer tonight(wednesday) at the weekly podcast and remind him he’s really missing out on this! 😛

    VMworld Day 1

    · ·

    It’s been chaos all day long yesterday. Some of you might have expected me to do actually live blogging and that’s what I had originally planned indeed. But after being added to the SRM Lab and last minute Community Lounge there’s no way I can find the time to do some actual blogging, or even twitter these days.

    I’m not even going to try to recap yesterday in terms of announcements. No way I can keep up with the constant flow of information/announcements being published / exposed by my fellow bloggers. Most of them are featured on PlanetV12N or use twitter anyway so you can easily keep track of what’s happening at VMworld.

    I do want to tell you guys about the VMTN Experts Session we had yesterday. Although the turn up wasn’t what we hoped for I do think the session was useful. All the VMTN experts and important bloggers, and I guess we can openly call them vExperts today, were there and you actually had the chance to talk to people who’s avatar you normally only see on Twitter or the forums. The conversations we had were awesome.

    I’ve got a totally different view of some people. Some of them I only know via their short tweets or replies on the forum or even blog articles. Some posts/tweets get interpreted a certain way because of a feeling/association you have with the person. Interpreting posts on the internet is hard and I know I will definitely think different about certain people and posts, which is a good thing.

    I was really honored to meet people like Mike Laverick, Scott Herold, Jason Boche and Tom Howarth. There contributions in terms of VMTN community posts and moderation, blog articles, books and just VMware / Virtualization evangelism in general is outstanding and I highly respect their input, effort and knowledge! Great to finally meet you guys!

    Now head over to PlanetV12n and read up. I will give my take on all the announcements when the storm is over though!

    PS: A reminder to some of the bloggers out there, if you visit the Live Labs and want to take pictures / film the technology preview please ask the VMware Employees if you’re allowed to do so…