DRS not taking CPU Ready Time in to account? Need your help!

9 May, 2013 By 15 Comments

For years these rumors have been floating around that DRS does not take CPU Ready Time (%RDY) in to account when it comes load balancing the virtual infrastructure. Fact is that %RDY has always been a part of the DRS algorithm but not as a first class citizen but as part of CPU Demand, which is a combination of various metrics but includes %RDY. Still, one might ask why %RDY is not a first class citizen.

There is a good reason though that %RDY isn’t, just think about what DRS is and does and how it actually goes about balancing out the environment, trying to please all virtual machines. Yes a lot of possibilities indeed to move virtual machines around in a cluster. So you can imagine that it is is really complex (and expensive) to calculate what the possible impact is after a virtual machine has been migrated “from a host” or “to a host” for all of the first class citizen metrics.

Now, for a long time the DRS engineering team has been looking for situations in the field where a cluster is balanced according to DRS but there are still virtual machines experiencing performance problems due to high %RDY. The DRS team really wants to fix this problem or bust the myth – what they need is hard data. In other words, vc-support bundles from vCenter and vm-support bundles from all hosts with high ready times. So far, no one has been able to provide these logs / cold hard facts.

If you see this scenario in your environment regularly please let me know. I will personally get you in touch with our DRS engineering team and they will look at your environment and try to solve this problem once and for all. We need YOU!

Filed Under: Various Tagged With: , , , ,

Tested / Supported / Certified by VMware? (caching / dr solutions)

8 May, 2013 By 5 Comments

Lately I have been receiving more and more questions around support for specific “hypervisor side” solutions. With that meaning, how VMware deals with solutions which are installed within the hypervisor. I have always found it very difficult to dig up details around this both externally and internally. I figured it was time to try to make things a bit more clear, if possible at all.

For VMware Technology Partners there are various programs they can join. Some of the programs include a rigid VMware test/certification process which results in being listed on the VMware Compatibility Guide (VCG). You can find those which are officially certified on our VMware Compatibility Guide here, just type the name of the solution in the search bar. For instance when I type in “Atlantis” I get a link to the Atlantis ILIO page and can see which version of ILIO is supported today with which version of vSphere. Note that in this case on vSphere 4.x is listed, but Atlantis assured me that this will be updated to include vSphere 5.x soon.

Then there are the Partner Verified and Supported Product (PVSP) solutions. These are typically solutions that do not fit the VCG, for instance when it is new type of solution and there is no certification process yet. Now of course there are still strict guidelines for these solutions to be listed. For instance, your solution will only be listed on the PVSP (and the VCG for that matter) when you are using public APIs. An example for instance is the Riverbed Steelhead appliance, it follows all of the guidelines and is listed on the PVSP as such. You can find all the solutions which are part of the PVSP program here.

Finally there is the VMware Solutions Exchange section on vmware.com. This is where you will find most other solutions… Solutions which are not officially tested/certified (part of the VCG) or part of the PVSP program because of various reasons. Note that these solutions, although listed, are not supported by VMware in anyway. Now, of course VMware Support typically will do its best to help a customer out. However, it is not uncommon to be asked to reproduce the problem on an environment which does not have that solution installed so that it can be determined what is causing the issue and who is best equipped to help solving the issue.

I am not saying that those solution that are not listed on the VCG or PVSP should be avoided. They could very well solve that problem you have, or be the solution to fulfill your business requirements and as such be the “must use” component in your stack. It should be noted though that when introducing any 3rd party solution that there is a “risk” associated with it. From an architectural and operational perspective it is heavily recommended to validate what that risk exactly is. How you can minimize that risk? What you will need to do to get the right level of support? And ultimately, which company is responsible for which part? As when push comes to shove, you don’t want to be that person spending hours on the phone just figuring out who is supporting what! You just want to be on the phone to solve the problem right?!

I hope this helps some of you out there who asked me this question.

** Note: the above is not an official VMware Support statement or a VMware Partner Alliances statement, these are my observations made while digging through the links on vmware.com **

Filed Under: Various Tagged With: , , ,

vOpenData – feed up!

15 April, 2013 By 10 Comments

A couple of weeks ago I asked this question on twitter about what the average disk size of a virtual machine is these days. Within a couple of minutes Ben Thomas replied and said we might be able to create a survey script and he copied William Lam in. Now for those who never worked with William, if you ask him a question like that you can expect him to knock something out… William and Ben decided not to just knock out a survey script, but rather an open community project called vOpenData.

vOpenData

This open community project consists of a script that collects the data (and they collect a significant amount, you can see what they collect here.) and is aiming to provide various trending statistics and data for virtualized environments. The data is fed back in to the vOpenData database. The vOpenData website has a great dashboard which provides you all these cool stats. For instance, at the moment there are 77 infrastructures that provided data to their collection. The question I asked, what is the average disk size, currently says “61.51GB”. That average is based on those 77 infrastructures with over 27.000 VMs in total combined. Nice right!?!

I have already emailed William a bunch of suggestions, and as I will be in Palo Alto this week I am sure some more will bubble up during conversations. I am hoping that everyone sees the power of a solution like this and can help feeding data in to the vOpenData platform.

Go here to download the bits and feed up!

** I have had some people asking me how vOpenData compares to CloudPhysics. I have also seen some people comparing vOpenData to CloudPhysics… To be honest you can’t really compare them. Where vOpenData is about averages and statistics, CloudPhysics is more about analytics and simulation models. **

Filed Under: Management & Automation, Various Tagged With:

Awesome Fling: vCenter 5.1 Pre-Install Check

22 March, 2013 By 4 Comments

One of the things that many people have asked me is how they could check if their environment was meeting the requirements for an upgrade to 5.1. Until today I never really had a good answer for it but fortunately that has changed. Alan Renouf has spent countless of hours developing a script that validated your environment and assesses if it is ready for an upgrade to vSphere 5.1.

This is a PowerShell script written to help customers validate their environment and assess if it is ready for a 5.1.x upgrade. The script checks against known misconfiguration and issues raised with VMware Support. This script checks the Windows Server and Active Directory configuration and provides an on screen report of known issues or configuration issues, the script also provides a text report which can help with further trouble shooting.

Is that helpful or what? Instead of going through the motion your just run this pre-flight script and it will tell you if you are good to go or not, or if changes are required. If you are planning an upgrade or are about to upgrade make sure to run this script.

Awesome job Alan, lets keep these coming!

Filed Under: Server, Various Tagged With: , , , ,

Startup Intro: AetherStore

22 March, 2013 By 3 Comments

AetherStore Every once in a while you see a solution by a startup and you get all excited. AetherStore is one of those type of solutions. The funny thing is that AetherStore is not directly related to my day-to-day job, but I can fully relate to their pitch. So what is AetherStore and who are the folks behind it?

AetherStore was founded by three graduates from Scotland’s University of St Andrews. This by itself is worth mentioning in my opinion as especially in this space I don’t typically see an enormous amount of innovation coming out of Europe. (Although since then they moved to the US.) With experience in distributed systems, fault tolerance, databases and storage it is not surprising to see what problems they are trying to solve and how they are intending to solve it.

AetherStore is indeed a storage solution as you probably had already guessed. AetherStore is all about using spare resources, and in this case storage resources. Essentially what AetherStore is aiming to do for your company is leveraging the available local disk space of your desktops (and servers for that matter) and offer that up as a “data store”. In other words; if you have 20 desktops with a 1 TB disk but only 100GB is used then 900GB of that disk can be used for other purposes. Now reality of course is that it isn’t possible to use the full 900GB for other purposes but you get my point.

AetherStore essentially is a distributed data store solution. This distributed data store is served up to users as a regular network file share and all the magic AetherStore does is hidden from the user. I guess the big question that pops-up immediately is what about availability, security and performance? All three of those are typically what either keeps the user, or the administrator busy. AetherStore solves those problems in various ways:

  • Performance: a local cache is used to optimize the end-user experience
  • Availability: Data is replicated to multiple “nodes” meaning that if a “node” fails than data can be reconstructred. On top of that AetherStore offers the ability to backup (and restore) data to the “cloud” (Mozy, Amazon etc)
  • Security: Data is encrypted

That is not all, on top of that AetherStore offers versioning of files and ensure efficiency by offering deduplication. I guess it all sounds very promising right? In my opinion it does, and it is one of those solutions that I have on my “watch lists”.

I do wonder what the requirements are when it comes down to availability of data when people move around different desktops; and desktops are also powered-off or restarted by users at random. I also would like to point out here that I have not played with AetherStore, neither is this article sponsored or am I affiliated with AetherStore in any way. This is simply and introduction to a cool startup which managed to intrigue / interest me with their technology.

If you want to find out more about AetherStore, make sure to sign up on http://www.aetherstore.com/ for early access if you are interested, and/or follow them on twitter. If you want to know more, I can recommend this white paper about AetherStore as it reveals some more of details of the implementation.

Filed Under: Desktop, Storage, Various Tagged With: , , ,

Startup Intro: SoftNAS

19 March, 2013 By 8 Comments

Last week I had a chat with Rick Braddy from SoftNAS. Some of you might know Rick from when he was the CTO of a hosted virtul desktop company called Virtual-Q and others from when he was the CTO of Citrix for XenApp and XenDesktop. Today Rick is the CTO for SoftNAS, a software and appliance based storage solution. Rick gave me an introduction to what it is SoftNAS (Professional) does and offers and I figured I would do a short write-up as an introduction to SoftNAS.

Ultimately SoftNAS is a virtual appliance that offers up local storage as shared storage. SoftNAS is build on top of CentOS and leverages ZFS. It is deployed as a virtual machine, which means that it takes a couple of minutes to set up. SoftNAS has a nice looking user interface which allows you to quickly create shared storage for your virtual environment. When I say quickly I mean in a matter of minutes you have shared storage to your disposal: select your volumes –> create a storage pool –> create a volume –> use it. For those who care, besides VMware vSphere SoftNAS also supports Hyper-V and Amazon EC2. [Read more...]

Filed Under: Storage, Various Tagged With: , , ,

VMworld 2013 call for papers open!

13 March, 2013 By 1 Comment

VMworld 2013 Call For Papers opened today. I know many of you are excited about this and hoping to have a session accepted at this top-of-the-bill event. Personally I always submit various sessions and for the last years always had atleast one of them excepted, and some years even multiple. So the obvious question I always get is “do you have tips”?

I guess at the risk of lowering my own chances, here they are, do note that as always there is a limited number of sessions that will be accepted. No guarantee that my tips will help in any shape or form, but these are my rules when I submit a session:

  • Be Original, make sure your suggestion is unique. If there are 10 others submitting the exact same, chances are slim you will get in. (Every year I see at least 5 or 6 community/bloggers panel sessions being submitted, with limited space make sure you work together on this and do not compete.)
  • Think Big, although your 2 host home-lab might be interesting to you VMworld is attended by 20.000+ people. So make sure your session appeals to the broader audience AND to the voting committee.
  • Quality over Quantity, really there is no point in submitting 10 sessions. Rather submit two or three, well thought out and developed submissions. This is what you will be judged on by the voting committee. They will read your title and summary, this will need to convince them!
  • Co-Present! Something I always try to do is to team up with someone. Not only will it make the “delivery” a lot easier, but a second pair of eyes on the submission will help!
  • Theme – Theme – Theme, although the theme for 2013 hasn’t been announced yet all of us can more or less guess what it will be about. We all know VMware has these three pillars they are focusing on, and that the SDDC is a major part of it… Make sure to submit something that hits that theme / fits in to those three pillars!

Now before you start running and submit a session, take a deep breath, sit down, relax and think about what you are going to submit. The Call For Papers will open until April 12th so no need to rush, start scribbling down those ideas and go take your time to fine-tune them!

http://www.vmworld.com/community/conference/cfp

Filed Under: Various Tagged With: ,
« Older Posts