Thoughts on Virtual Infrastructure Management

Manageability is Top Server Virtualization Concern

By Lisa Crewe

According to The Info Pro’s Wave 6 server study, manageability and performance under load are the top server virtualization concerns for IT pros at the Fortune 1000 and mid market enterprises they interviewed. 

Manageability and Performance Under Load are Top Virtualization Concerns

Manageability and Performance Under Load are Top Virtualization Concerns

 This isn’t surprising and really validates what Akorri has been saying for the last few years.   And that is, that you need a virtual infrastructure management tool focused on performance and resource utilization to manage both your virtual and physical infrastructure holistically. We all know VM sprawl is a real issue.  TheInfoPro is coming out with their Server Wave 7 report shortly and we have the pleasure of hosting an upcoming webinar featuring Bob Gill, TheInfoPro’s Server Research Director sharing key findings and advice to help you manage your virtualization deployments more effectively.  You can register here.  If  there are any issues you’d like addressed during the webcast, please contact me.

March 31, 2009 | Filed Under Education, Virtual infrastructure management | Leave a Comment 

Living in the IT Petri Dish

By Rob Strechay

It occurs to me that problems in an IT  infrastructure are like germs spreading in the Petri dish that is a kindergarten class.  Germs grow and grow, kids get the sniffles, and then the entire class gets sick.  Ultimately, the illness spreads to everyone in the school and at home (yes I have young children).  The same thing happens with applications that share infrastructure.  One server could be operating in an unhealthy manner, which can, and usually does, impact its neighbor.  This “spreading of germs” is only made more violent and unpredictable with the advent of server virtualization and shared storage.

It only takes one “unhealthy” vm to spoil a cluster or one poorly performing vm to start others pin-wheeling around based on simple thresholds.

How do you determine the root cause?  If you are simply looking at cpu, memory, or disk utilization that is equivalent to waiting for the kid in the class to have hives and 102 fever before you start to take action.  And when you do move the kid, how many more are already “unhealthy”.  But in a vm world we move the unhealthy vm to another server which would be equivalent to keeping the sick child in the classroom for the rest of the day.  So many more might become “unhealthy” or at the very least annoyed … sound like the day a few people called and said “my application is slow today?” 

 If it doesn’t fit there we move it again, to where it might fit more nicely.  Isn’t the definition of insanity doing the same thing over and over again?  Couldn’t we see that if the kid is complaining maybe he should get checked out and fixed?  OK, enough of the analogy. VMs don’t complain, end-users do.  So how do you know when workloads shift?  What is a normal workload?  What workloads play nicely together?  Is it a heavy cpu, memory, or IO workload.  What combination is right?  Is the vm doing un-natural acts … servicing more workload than it was spec’ed too?  More on that next time.

March 23, 2009 | Filed Under Performance, Virtual infrastructure management | 2 Comments 

Virtual Data Centers: The Next Five Years

By Lisa Crewe

I recently got to work with Gartner Research Vice President George Weiss, who was being interviewed on the evolution of virtualization and wanted to share his predictions for the virtual data center over the next five years.  In a nutshell, he says:

  • Organizations will have to engage in resource planning and optimization
  • Virtual machine densities will increase
  • Business critical “production” applications will be virtualized
  • Data center complexity will increase
  • Heterogeneous platforms will coexist

I think his predictions are right on.  For example, Microsoft just recently said it’s OK to virtualize SQL server, so I bet you’ll see more people trying it. In general, IT shops are trying to get more out of their virtual infrastructure, so they’re going to have to virtualize more.  At Akorri, we work with many organizations that are what I would call, advanced virtualization users, and we’re already seeing some of this stuff.  Some of our customers are scaling vm densities from 20 or 30 to 1 to 50 or 60 to 1.  Of course, you can’t do that without figuring out how it will impact performance. 

Check out George’s interview in an on demand webcast. It also includes an interview with Mike Moran, Senior IT Manager at Kronos talking about how he’s championed virtualization in his company, the challenges he’s faced along the way and how he’s now managing it.  I’d love to hear your thoughts on George’s predictions and find out which production applications you’re virtualizing.

March 18, 2009 | Filed Under Customer Experiences, Industry News & Opinion | Leave a Comment 

Virtualization Version 2 and Key Performance Indicators That Put the Science Back Into IT

By Rob Strechay

As IT professionals continue to take on more and more responsibilities, consistently we hear the pleas, “I need you to tell me what’s good and what’s bad” in my environment.  People are really looking for the answer to the question, “so what?”  They have a lot of stats coming at them from different directions, but they still ultimately rely on guesswork to make decisions. 

As we move into this virtualization version 2, or what I am coining VV2 … something like WWII … Key Performance Indicators (or KPIs) are going to be a must.  These are things that we may not fully understand the math behind, we just know that when it’s low people don’t complain and when it’s high they do.  Akorri’s Performance Index (PI) is just one of them.  It tells you what combo of cpu, memory, and IOPS is healthy or not given the infrastructure allocated to the workload.  What this reminds me of is the “miles till empty” message most cars have today.  The ultimate KPI is the “miles till empty” that is somehow doing math based on the speed and distance you are traveling to indicate if you are totally screwed or if you are going to make it to the next gas station.

I think of our PI as being that “miles to empty” for IT folks in a virtualized environment.  It provides the answers to the “so what question” … so will I be okay if a host fails or not … so what subscription ratio can I really get to … so is there too much workload on a virtual infrastructure?  Not an exhaustive list, but a nice set of questions to have answered, right?

Companies relying simply on  intuition, a hand full of data points, and qualitative “I think so” are in for a long year.  Answering the above questions with quantitative information will be part of the answer for companies to utilize IT as a way to weather the storm that is the current economy.  KPIs will really be a necessary part of putting the science back into IT, because right now it is far too many times an art and no science.

March 6, 2009 | Filed Under Customer Experiences, Performance, Virtual infrastructure management | Leave a Comment 

Customer Support Matters

By Rich Corley

During the past year we have invested a great deal of time and effort in developing Akorri’s software maintenance and support capability.  There’s no question that support is one of the big differentiators between software companies that are loved by their customers and partners, and those that are not.  To quote Steve Jobs, one of our company goals has been to build an “insanely great” customer support capability, and we’re working really hard on it.

Our objective at Akorri has been to try to build a business that is innovative on multiple dimensions.  The top three dimensions Akorri is focusing on are (1) product, (2) channel, and (3) customer support.  We believe BalancePoint is an innovative product and are continuing to focus engineering on innovation while meeting customer and market needs.  With regard to the channel, my goal has been to become recognized by channel partners and their customers as their favorite company to work with in the virtualization management space.  With regard to the customer support dimension, I believe we have made excellent progress.  Every week now I get emails out of the blue from users and potential customers about how one of our support people (Paul, Robert, Don, James, Fritz, Rick, Mike, John, etc.) went the extra mile.  I love that! 

In terms of how we are trying to innovate in support, I don’t mean to suggest that our support offering is entirely unique, or perfect (yet), but there are a number of things we do that I think stand out:

  • When a potential customer evaluates our product, we provide them full access to our customer support and treat them exactly as if they were a paying customer.  Many companies don’t do this because it costs money, but the way we see it, you need to have a chance to evaluate not just the product but also the support.  Our support is available 24×7x365, and can be contacted by phone, email, through the support portal, or at www.akorri.com/support.
  • When we come out with new releases, upgrades have been covered under support, for both minor and also major releases.  When we came out with BalancePoint 1.5,1.7, 2.0, and 2.3 all our customers got upgraded.  Our next major release will be 3.0, and all our existing customers are going to get that without an additional charge.  Other software companies also do this, but many charge for major releases.  We have not done that.  As with 2.0, in 3.0 we will be offering major new functionality, and that is included under the support agreement. 
  • Customers get a login to our customer support portal and a very comprehensive knowledge-base, technical FAQs, documentation, updates, etc.  We also have online, on-demand training and certification.  Again, evaluation users get access to the support portal and get a login to the online training site during their eval period.
  • We do the installation for you, without an additional charge.  You can do the install yourself if you want, but we are happy to install the software.  We do this remotely from our headquarters.

Gartner Group recommends that a software company’s support capability be evaluated on four criteria: (1) installation support, (2) ongoing support, (3) availability, and (4) support reputation.  We are doing our best to build a support capability that exceeds expectations on all of these criteria.  Let me know how you think we’re doing at rich@akorri.com.

March 4, 2009 | Filed Under Customer Experiences, Inside Akorri | Leave a Comment