SandRidge deployed server virtualization over two years ago and has virtualized 60% of its application environment (including test and production applications). Today, over 140 machines are running on nine physical hosts supported by NetApp storage. The complex environment made it challenging to troubleshoot performance problems using the existing element management tools. When performance problems occurred, SandRidge consulted four different tools to resolve issues, which took a few systems engineers and up to a day to review data and find the “needle in the haystack.”

Cameron Han is the Infrastructure Operations Supervisor at SandRidge Energy, Inc.  I’m thrilled to share that he’s presenting his story at VMworld on how SandRidge’s IT department was able to improve their infrastructure performance using Akorri BalancePoint.

Here’s what he had to say about the product.

“BalancePoint is the only tool we found that can decipher our virtual machines, storage, and operating systems together, and we like that it will proactively notify us of potential performance bottlenecks.”

Check out the Session @ VMworld

Who: Cameron Han, Senior Systems Engineer at SandRidge Energy, Inc.

What: SandRidge Energy Increases IT Virtualization Efficiency

When: August 31, 2010 at 3:30 p.m.

Where: VMworld 2010, Moscone Center, San Francisco

Session ID: V18113, Track: Virtualization 101

If you can’t make the session, you can read the case study here.

See Us at VMworld, Booth #1331

• See a demo of BalancePoint v3.5 that Predicts Performance Issues.
• Get a cool t-shirt and wear it for a chance to win a new Apple iPad or GPS.
• Get a free drink (or two or three) during the VMworld “Hall Crawl”, Tuesday, August 31 from 4-6 p.m.

Stop by the Akorri booth and visit us!  Tell me you read my blog for an additional chance to win.

I predict HP will buy 3PAR.  It could happen.  Or, Dell could outbid HP as Dell is expected to make another run at the prize.  If I knew the answer, I would tell you.  

What I do know how to predict is infrastructure performance issues.  (You had to see that one coming.)

Along with a bunch of our friends in the virtualization management space making pre-VMworld announcements, Akorri introduced today our newest version of BalancePoint.  

The most exciting new capability is the ability to prevent outages and bottlenecks in a virtual infrastructure. The first in a series of predictive capabilities, BalancePoint Predictor™ for Storage automatically detects issues that can quickly lead to application outages.

No other product has this capability.  BalancePoint looks at the workloads, capacity and a new analytic, disk utilization, and what pops out is an email that tells you the causes of the problems. 

 The “bullies” are defined as the potential causes.  They are the VM workloads competing for disk resources. 

It also identifies the business impact or “victims” which are the workloads that are experiencing slower response times.  All of this is delivered in an email that gives you direct links into BalancePoint to investigate further. The information can also be sent as an alert to a 3rd party tool.

We think it will save anyone administering a virtual environment a ton of time and aggravation.  You can learn more in the press release and by visiting Akorri at VMworld, booth #1331, to see a demo.   We’re not giving away $5k, but we do have free beer (on Tuesday from 4-6) and iPads and GPS’s to giveaway, so I hope you’ll check us out.  Hope to see you there!

Just learned about the VMworld Fun Run today and I am really psyched.  I happen to be on a running kick this year having finished my first sprint triathlon and I’m psyched for a reason to bring my running sneakers to San Fran. In all honesty, I would have brought them anyway, but this is more fun.   Who wants to race?

Now for the “Now What” part. This is trickier to get your hands on. The challenges of getting to the next level of ongoing savings beyond just server consolidation include:

• Achieving higher VM densities (more than 10 ESX’s and more than 100 Virtual Machines) and/or virtualizing Tier 1 applications
• Understanding  how this impacts the rest of the data center resources.

Adding Storage to the Equation

While server savings are very visible, they could be getting offset by the lack of visibility and connection to the storage environment that is critical in supporting production applications for important business operations. If storage resources are over provisioned or are not tiered or configured correctly to the proper performance for the VM’s  they support, then the costs associated with that is going to offset the server savings we got with virtualizing to begin with.  This is going backwards and begins to erode saving capital and management dollars. How can we stop that and be sure that in managing these new virtualized pools of resources, we have the end to end visibility, capacity, performance information, reporting and insightful analytics to make the right decisions and avoid costly outage issues as well as insure we have the efficiencies promised by virtual technologies.

The second wave of ROI for virtual data center is about taking the assets you have deployed (server and storage) and the operations management  talent you have and making them super efficient. Key considerations are:

1)      Maintaining High Service Levels  & Reduced Troubleshooting Time

Time wasted trying to diagnose and troubleshoot an application outage that is being blamed on the data center. Here it’s guilty until proven innocent and each group (server, network and storage groups) must try and gather, compile and assimilate abstracted dynamic data and pinpoint a cause for remediation.  If this is taking longer than in the physical world to figure out, that is a cost of resources we never get back and subtracts for the initial savings.

2)      Reducing Infrastructure Costs  with Better Capacity Planning

Where in the infrastructure is there additional capacity that not only meets the simple capacity test, but also meets the application I/O work load needs of that application group. If I try and add more VM’s to an ESX that looks like it can handle more, then why is the app running slower than before? If we don’t integrate the storage capacity and performance characteristics, then we are making the VM density better at the ESX level, but the business is screaming that the performance got considerable worse. More time wasted trying to figure out why that happened and you probably will stop attempting to VM density stack anymore. What are the costs associated with your top performing apps when they aren’t running or running at very slow rate?

Validate before you buy. If you are about to make a significant cap ex purchase, be sure you don’t have available capacity in the infrastructure and that it is going to meet the needs of your performance problem or new app roll out. Do I need Solid State disks or does the history from the application I/O profile suggest you don’t need that performance?  Are you sure it’s an ESX memory issue versus a storage disk contention problem? The only way to understand that is to look at the integrated I/O supporting that application and measure and track that across the infrastructure supporting the transaction. What if you had the ability to drive cap ex savings by even a 5% to 10% utilization of the existing investment you have?  This alone drives significant savings and cost avoidance/delays freeing budget up for more pressing projects.

3)      Increased Productivity

How much time and effort can be saved by having data center wide dynamic reporting in place to meet operating and management requirements and the dashboards that give the information needed to respond to problems quickly so that they don’t linger?  Can you predict when performance is trending down and capacity is trending towards an outage? Taking the guesswork out of any operations task allows you to regain control and then proactively manage this new dynamic operations environment at the lowest cost of service possible.

Finally, having a solution that provides you  the same metrics and capabilities for the physical environment as the virtual environment saves you money in time and better utilization. 

 Calculate Your Potential Savings

Check out this easy to use ROI calculator, for getting the next wave of infrastructure savings off and running.

• Myth #1: Virtualisation guarantees performance
• Myth #2: Virtualised IT silos can be managed in isolation
• Myth #3: Virtualised IT can be managed separately from the physical infrastructure

You can read the article here.  What other myths would you add to this list?

1. You bought more memory for your ESX server and it didn’t fix the performance problem.
2. VMware DRS causes you performance problems every time it moves a virtual machine.
3. VM performance problems come and go and you can’t figure out why.
4. You haven’t turned on VMware DRS because you’re afraid of what it will do to your performance.
5. The SSDs you purchased are not delivering the performance you expected.
6. You ran into poor performance due to running a VM on a LUN with the wrong RAID level.
7. The new array you purchased doesn’t perform as expected in your virtual environment
8. You spent days debugging a VMware related issue using data from several different tools.
9. Your application owners complain that their application performance on VMware is worse than when it was run on a dedicated physical server.
10. Application performance problems are experienced by end-users and you found it’s not the network or the application.

Do you have one we didn’t mention?   I’d love to hear about it and add it to the list.

Why not just use something that people understand?  Well, because it’s different.  BalancePoint often gets confused with traditional storage resource management products or virtual machine management tools. 

What’s a marketer to do?  Well, what we decided to do is create this video that uses an analogy to tell the “how BalancePoint is different” story.  

Our story begins…If you’re the Admin managing a virtualized data center, every morning can be like driving into a new city…

 Watch the video and let me know how we did.

According to Forrester, capacity management is the top operational concern among IT professionals using virtualization.  IT Business Edge recently published an article I wrote on this topic that I thought I’d share titled, Assuring Performance and Capacity Management in a Virtual World.

You can get a handle on it and there are some tools out there to help.  Read this article to learn how.

If you’re managing performance of the virtual machines and the physical machines separately it can make you feel like you have two jobs.  Who has time for that? 

What’s needed is a solution that handles both and can provide the following:

• A metric that proves your VMs always perform optimally using minimal server and storage resources.
• Measure and record the round trip response time of your virtual infrastructure? (virtual/physical; storage/server)
• Alarms that isolate specific applications that have poor performance due to disk contention
• A topology map that automatically discovers and displays I/O paths through virtual and physical servers and storage

Learn how you can get that combination.  Watch this video.