Irfan Ahmad presenting, he presented last year on what became Storage I/O Control.
- Storage DRS is a “stealth” project at VMware
- Big problem is a VM admin doesn’t necessarily know what class of disk or how many spindles are behind a particular datastore.
- Create a new primitive called a “datastore group” which is a new domain like a DRS cluster. Note from me: This will of course dovetail nicely into vCD service levels!
- Storage DRS would automatically load balance across multiple datastores in a datastore pool.
- When you create a new VM, you place it on a datastore group and it does auto-placement. Takes both free space and I/O into consideration
- ESX host will gather both free space and I/O stats to help balance for initial placement as well as Storage VMotions.
- Cluster-datastore group relationship is many to many – a datastore group can span clusters and a cluster can have multiple datastore groups.
- Recommend all datastore groups are visible from all hosts but not enforced, much like existing datastore presentation in clusters. Storage DRS will do best effort if you don’t follow this, though.
- You can have Storage DRS affinity rules – keep these two VMs on different arrays or keep all disks from this VM on the same datastore.
- Datastore Maintenance Mode! Say this datastore is going down and it’ll auto SVMotion all VMs off into other LUNs in the same pool.
- You can of course add datastores to an existing group like adding a host to a cluster.
- When you enable it you can do it on Capacity or Capacity and I/O (important for virtualized arrays like EVA where a group of LUNs shares same performance pool)
- You can set an I/O latency metric so that if latency gets above 15ms it’ll move stuff to a datastore with lower latency. Really smart way to determine if I/O capacity is too high.
- Balancing will only happen every few days, not every hour like a VMotion.
- Initial placement will take into account both DRS and StorageDRS metrics, as well as how well connected the datastore is. For example, if datastore is hooked up to all hosts it will prefer that over a datastore only connected to 1 or 2.
- You can balance capacity based on keeping them all withing a % of each other (in other words prefer balance) or just try to keep them below a certain % (just try to avoid making a datastore full)
- It even takes into account growth rate of thinly provisioned disks when determining a good placement! Wow, that’s smart. Weights powered-off VMs less since their I/O is generally 0 when off.
- Prefers moving VMs with low SvMotion overhead (like move smaller VMs before big ones, like DRS).
- Does load balancing by knowing a more powerful datastore (one with more disks behind it) will have latency degrade slower than a less powerful datastore. This insight is used to model the performance to make smart migration choices.
- Also model metrics of individual virtual disks and feeds that into the model.
- They did a man vs. machine test. Made 13 VMs with standard workloads, gave all info to 2 storage admins at VMware vs. Storage DRS algorithm. While IOPS were about the same between experts and algorithm, algorithm beat them significantly on latency!
This was the best session I went to all day. This once again reminds me why VMware makes such cool products – they really understand problems and they have really smart people trying to solve them.
Cool highlights from this session, I’m going to skip all the pre-ThinApp 4.6 stuff since it’s old news:
- ThinApp (as of 4.6) now does transparent page sharing for applications (like in TS environments) just like ESX does for VMs. Pretty neat!
- In ThinApp 4.6 you can “harvest” IE6 straight out of an XP system and run it on Win7. Only WinXP specific file that gets put in the package is Shell32.dll, otherwise icons and menus don’t work correctly. The resulting ThinApp performs like IE6 SP1.
- ThinApp Converter which is included in ThinApp 4.6 allows you to automatically build a package from an automated installer. You can take existing install packages (like from Wise, Alritis, or LANDesk) and convert them to ThinApp packages easily with a simple command line and a blank VM.
- The futures stuff was talking about a new features called “ThinApp Factory.” It’s a prototype which can download RSS feeds of applications and automatically download the app installer, silently install it and capture the package using ThinApp Converter, and then publish it to users or allow users to download it from an “App Store” kind of thing. This will probably feed into the Horizon stuff they showed this morning in the keynote.
Session is around updating the reference architecture for View 4.0 to View 4.5.
- Goal of View is to delivery Consumer Cloud experience for the Enterprise.
- Goal of the 4.0 reference architecture was to simulate a realistic desktop workload, validate 2,048 users.
- Session turning into a pitch for UCS very quickly…
- Now they’re off to talk about RAWC, which is really old news. New version of RAWC supports simulating workloads on Win7. Still can only simulate a preselected set of apps, no custom app load testing. You can learn more about rock on its Youtube channel.
- View 4 reference architecture was run on UCS, CX4, vSphere 4.0 and WinXP SP3.
- Just released – Win7 optimization guide, includes a BAT file that optimizes the VM for you! Already found it here.
- Going through all the stuff they had to do to the Storage to make it perform. Wouldn’t it have been nice if it was on virtualized storage and you didn’t have to worry about RAID groups and all that crap? :)
- View 4.0 Reference architecture they got up to 16VMs/core. I think this is super aggressive and I don’t recommend customers size for this #.
- Finally we get to View 4.5 stuff! Talking about the new Tiered storage capabilities of View 4.5
- They’re putting SSDs in each physical server… again more Cisco specific stuff. I think sticking SSDs in every server drives up the cost too much. Plus wouldn’t it kill vMotion.
- They did the View 4.5 test with a single non-persistent pool.
- They see CPU being a bottleneck on optimized Win7 32b deployments… but they were only giving each Win7 VM 1GB of RAM.
- During the Q&A, asked about HA/VMotion. This reference architecture doesn’t allow for VMotion or HA. And Non-Persistent pools require some sort of 3rd party profile management to make it work. If you want to take a system down you’ll have to do it after hours. Don’t like it! I’ll stick with SANs to give full functionality instead of neutering 1/2 the Enterprise functionality.
Something amusing that I noticed in the compatibility matrix a couple weeks ago:
Holy cow, they finally got VMware Server support in vCenter! It only took them… 4 years? I wonder why they even bothered. Anyone wanna try it out?
I”m here in San Francisco and I will be blogging and tweeting about VMworld! Follow along on twitter (I’m @justinemerson) or check the recent tweets on the right.
Sunday and Monday I’ll be busy with PTAB meetings (Partner Technical Advisory Board) doing stuff I can’t tell any of you about. =) But Tuesday through Thursday expect content from here. Hope to see you all there! I’m the nerdy looking redhead. Come up and say hi!
Greeting everyone. Been very busy preparing for a technical conference my company is putting on, but I thought I’d post on one of the issue’s I’ve come across in the lab.
When upgrading a vCenter 4 server to vCenter 4.1, you aren’t given the option of selecting the LocalSystem account because the installer forces it to be the currently logged-on account. This broke my first upgrade of vCenter and I had to roll back.
There have been some recommendations on the forum to simply change the account type back to LocalSystem afterwards, however I thought my solution was slightly more elegant: run the installer as LocalSystem. This is more difficult than it sounds. In each MS release Microsoft has made it a bit harder to do this (the trick using the “at” command doesn’t work anymore, nor does the Scheduled Task trick). The easiest way it turns out is to use PsTools from SysInternals. To launch the vCenter 4.1 installer as LocalSystem, run the following:
psexec -i -s D:\autorun.exe
(where autorun.exe is the installer shell on the vCenter CD)
Note you must run psexec from a UAC elevated prompt. You can substitute any command you want at the end there, but it’s recommended to launch the installers from the autorun screen.
One of the great things that came out of HP Tech Forum last week was the official announcement of HP’s various reference designs around their VDI solutions. The hub at HP’s website is here, as of right now only the VMware View architecture PDF is up but the XenDesktop ones are coming (one for XenServer, the other for Hyper-V). Some aspects of this reference design were announced all the way back at VMworld 2009 and are only now coming to fruition. This is mostly because of this bad boy:
These are the building blocks of the HP P4800 Blade-based SAN solution. HP took their SAN/IQ software and loaded it directly on blade servers, which then attach to external SAS-attached storage to create a 10GBit iSCSI SAN inside the blade chassis. No 10GBit core or switches required! The P4800 is designed specifically for VDI solutions and currently is only available for that kind of solution (although there’s nothing stopping you from using it as a general purpose iSCSI SAN, it’s not recommended because the I/O patterns for VDI and normal server workloads are very different).
This is HP’s flagship VDI design. Going forward there will be more reference designs for smaller deployments, going all the way down to Micro-Branch type deployments with just two servers and no dedicated shared storage but still full redundancy. All are based on the P4000 SAN.
So I’m not trying to make this an advertisement here (although I do think it’s really cool), the reason I’m linking to this is that HP has done a ton of validation and testing around the solution and have provided some great numbers around storage requirements per user for VDI environments. They’ve broken down users into Task, Productivity, and Knowledge workers. According to their numbers, on average these will take 6, 9, and 22 IOPS respectively to support. This can be very demanding on your SAN, and the #1 hardware design related problem users run into is sizing their VDI environments based on capacity and not on performance. These sizing guidelines should help anyone looking to architect a good VDI solution.
I don’t really have a very high opinion of Oracle as a company. Their products can be quite good, and have a great following, but as a company they are very unfriendly to their customer base. You can go back and look at their spats with EMC and VMware, or their ridiculous licensing terms for any virtual environment other than OVM. But their latest acquisition of Sun has certainly been an eye opener to how they treat their customers:
- You can no longer download drivers for their Sun servers unless you have a valid support contract (!!)
- Oracle has ended a lot of programs Sun used to run in the education space that kept them in the game. I know plenty of EDU customers who are now looking at other options because Oracle has just priced themselves out.
The latest chapter, outlined on The Register today, is that Oracle has terminated the agreement with HP to support Solaris x86 on HP ProLiant servers. This is a big deal for customers who are looking to migrate off of Sun hardware, or who are already using HP hardware to run Solaris. And that’s exactly why Oracle terminated the agreement – it was driving people away from SunFire servers and to the competition who made a better product. The linked article mentions that Dell still has an agreement in place, but I think the writing’s on the wall for the OEM agreements with Dell and IBM, as well.
Bottom line: If you’re a Solaris customer and you want to have a future where you’re not forced into buying Oracle servers, storage, OS, and application stack just to be supported, maybe it’s time to start looking at Linux alternatives.
So it was with quite a lot of surprise that I opened my email this weekend and saw that I have been awarded the title of vExpert for 2010. I find it almost ridiculous that I’m up there with the other folks that have this award. I won’t bother naming them because invariably I will leave someone out who deserves this award much more than I.
The award I guess is kind of a wake up call – as you can see my last update here was some time ago. I could give excuses about work schedule or vacations or whatever but really the onus is on my and hopefully in the coming months I will show that I was worthy of this honor. Thanks again for all of you who read and comment.
I’m here in Las Vegas this week at Microsoft Management Summit 2010. Check my twitter feed for more constant updates, I will be posting a wrap up of various things at the end of the week.
Got the high score last night at the SVT booth on Geometry Wars Retro Evolved 2, hoping to win an XBox by Thursday :)