We recently had the opportunity to listen to Charles Babcock, editor-at-large of InformationWeek, speak about the much discussed topic of cloud computing at the TeamQuest Technology Summit. Babcock reinforced a couple of themes that we have been saying all along during the advent of cloud computing.

Cloud computing is not a revolution, but as Mr. Babcock terms it, a “convergence” of several technologies used together in a unique way. When it comes right down to it, a cloud environment still consists of traditional compute fabric we all have grown accustom to in the past several years with virtualization in the forefront. The main twist to the technology of cloud computing is the method of delivery to the end user.

Listen to the Clip: Cloud Computing Evolution vs. Revolution

We need to avoid “compulsive over-provisioning,”  which is something that we have seen in a majority of IT organizations in early stages of Capacity Management Maturity. Joining the journey to becoming a mature IT organization takes time and effort and a key component is understanding the current state of capacity coupled with forecasted capacity requirements. Having strong capacity management processes in place allow you to focus on bringing value to the business versus reacting to performance problems and incidents that arise due to capacity issues.

Listen to the Clip: Compulsive Over-prosivisioning

I like the idea of using the cloud to dodge problems with over-provisioning. I should point out though, that contrary to popular belief, tapping cloud computing still requires careful performance and capacity management. To achieve IT Service Optimization using the cloud, you need to analyze the financial aspects as well as performance, capacity, and risk when determining how, how much, and when to use cloud services. Keep in mind, the cloud is more complicated than any one of the technologies that converge to comprise it. As you prepare for it, you’ll want to tap the right performance and capacity tools to ensure that cloud projects successfully deliver business value.

If you are interested in learning more from Mr. Babcock regarding cloud computing, check out his book, Management Strategies for the Cloud Revolution. You might also be interested in this white paper from TeamQuest: Capacity Management Ensures Success for Enterprise Cloud Consumers.

Engineer Mark Manness’ breakout session focused on understanding that performance instrumentation at all levels (e.g., application, guest OS, virtual machine, host) is necessary for successful capacity management on VMware. The three main takeaways from his presentation:

• Monitoring inside the Guest OS is an important piece of the performance puzzle
• What some of the Host and Virtual Machine metrics mean and how they can be used.
• Appreciate the numerous modeling options available on the VMware platform.
  • Physical to Virtual planning
  • Proper configuration of virtual machines
  • Using data from smaller benchmarks to size larger workloads.

Thanks for the post Mark.

TeamQuest engineer Scott Johnson wanted the audience to walk away with the following main points from his presentation:

• While measuring application response time can be a challenge, there are some reasonable options for passive data collection.
• A fortuitous side effect of some response time monitoring agents is additional data that can be used for application trouble shooting and business intelligence.
• Application measurements can be used to enhance capacity models.

Johnson also noted that transaction rate and response time information are very valuable for monitoring service level agreements. In the world of Unix, Linux, and Windows, there is no standard transaction monitor for applications. For those that grew up in the mainframe world with proprietary transaction monitors, this situation is a bit unsettling. However, several approaches have been developed to address certain classes of applications.One approach for web-based applications is to use synthetic transactions and response time “robots” to initiate and measure these transactions. This approach is called active measurement since it initiates a small artificial load on the application in order to obtain measurement data. This approach provides reasonable indicators of application availability and network latency.Another approach is to instrument the application and then make the response time information available, perhaps through an application log file. Also, the technology used in web applications provides the possibility for web server log file analysis and TCP packet sniffing. These last approaches are usually called passive monitoring since they add no transaction load to the application. The advantage of passive monitoring is the potential to understand the actual rate of application activity and the actual user experience.Thanks to Scott for this post.

The guy pictured above was a part of the entertainment at the North American TTS 2010, where this year’s theme was “Optimize Your IT Zoo.” The San Diego Zoo is famous throughout the world, so in addition to the technical content, there were a number of visitors present who work at the zoo, including the beautiful bird shown here.

In some cases the technical content of the show was blended together with the zoo-based theme. In fact, one of my favorite presentations at this year’s event was a talk by the zoo’s Chief Technical Officer, Robert Erhardt. Here are a few tidbits from my notes:

• Google Street View works inside the San Diego Zoo. Check it out!
• The zoo attaches tiny fraction-of-a-gram RFID devices with little whisker-like antennas to rare mice in order to track them.
• Condors have radios that send data via satellites. The trick needed to make the radios practical was to design antennae that did not interfere with the birds’ ability to fold their wings.
• In California, organizations can qualify for big rebates by implementing power-saving virtualization solutions. They measure your power consumption before and after virtualization and your rebate is calculated based on the reduction in power usage.

I thought Erhardt’s talk was both entertaining and informative. It was good fun that I think exemplifies this year’s summit. If you didn’t make it this year, don’t miss next year’s event which is sure to be even better. At next year’s summit we will celebrate TeamQuest’s 20th year of existence!

This survey was taken at the 2009 TeamQuest Technology Summit, April 28-29. We asked attendees to choose their highest priority for 2009. And the winner… Overwhelmingly, it’s virtualization. Is that a surprise? Well, no, but I have two questions. Why is cloud computing so low with a paltry 7 percent and when will processes such as ITIL finally get their comeuppance?

Virtualization - 68%
Virtualization has been the teacher’s pet for the last couple of years. The news articles, analysts and businesses continue to sing its praises. We’ve branched out though. Instead of focusing on server virtualization, we’ve moved to desktop and application virtualization. What’s next? How else are you exploiting virtualization capabilities? Will virtualization continue to be the class favorite?

ITIL - 14% 
Ahhh yes. The processes that work well for IT, but have yet to translate to or get buy-in from the business side. ITIL is a tricky one for me. I see its value. ITIL isn’t easy. ITIL can demand a cultural shift and buy-in from IT staff. ITIL has several processes with just as many starting points. What are you to do?

With the mantra “do more with less” being trumpeted about companies and IT staff, one would think more businesses would welcome some sort of process to improve the delivery of services. Those of you who have had success incorporating an ITIL process or two should share your secrets with other readers. How do you get buy in from the business side?

Green IT - 11%
It’s good to see Green computing working its way up as a priority, but there’s more we in IT can do. I’m talking more than server consolidation. Ron Potter has a few ideas for data center and systems management professionals.

Cloud Computing - 7%
Cloud computing has captured the interest of the IT world in much the same way as virtual servers. It’s a great conversation piece, but its priority seems to be pretty low. Why is that? Is it due to security reasons? Are organizations concerned with someone else being in control?

Craig

In a poll of IT professionals at the TeamQuest Technology Summit, 85 percent said they were either reactive or more inclined to be reactive in their IT environment.

Maybe it’s the nature of the beast. IT is under demand to produce “more with less” and deliver always-on services. Marketing has a great idea and wants to run a campaign. Are we sitting at the table when marketing pitches a resource-taxing promotion on the company’s infrastructure? Do we allow ourselves enough time to test? Are we given enough time to plan for the campaign?

Some say you can’t sell prevention. But you can sell success. By proving that IT must be part of the strategic leadership on revenue-generating items, the company should have fewer hiccups, decreased headaches and more opportunities to capture revenue.

We’re all after the same thing – success for the company and happy customers. IT, by being ready (i.e., proactive), can help improve a company’s chances by using proper processes to improve revenue-generating and customer-facing opportunities.

For those of you who have had success at being proactive, how do you do it? What needs to happen to move toward being more proactive? Is it better software, processes, or people? Is it a combination?

We want to hear from you.

Craig

Steve Henning briefed the TTS audience on the partnership between his company Integrien Corp and TeamQuest. Integrien offers advanced analytics as a solution to over-alerting by monitoring software. Typically, firms receive hundreds of alerts. One example he gave was of a company which received hundreds every hour. In such a climate, alerts eventually get completely ignored.

Integrien proved its value at one large customer by conducting an analysis of historical data and then laying out what its system would have advised concerning the IT emergency that had been experienced in that period. During that emergency, hundreds of alerts from its existing monitoring system had failed to zero in on the real issues. With Integrien, a handful of alerts highlighted the exact problems and would have warned IT up to three hours in advance of the actual crash.

Integrien has chosen to tightly integrate its software with TeamQuest in order to offer the best of capacity planning alongside intelligent and predictive alerts. Integrien is in the midst of rolling out its solution along with TeamQuest in a large financial institution.  Craig

At the TeamQuest Technology Summit, John Miecielica of Metavante gave an informative talk on how to combine performance management and capacity planning with monitoring. He explained how much easier it used to be to do performance testing when you only had one server attached to its own storage with a defined set of users. Nowadays with virtualization, it’s a lot tougher as multiple users and apps are sharing a common infrastructure.

He gave an interesting example of an application that was suddenly exhibiting performance issues, yet had no change in traffic patterns and was running the same workloads. The application and infrastructure staff insisted no changes had been made. Performance monitoring indicated a change in IO and how it was serviced, yet the SAN team said there had been no microcode changes, new apps or major changes in storage subsystem. The issue had been caused by another team that had loaded an application on a shared server.

They had been very happy with the success of their implementation and had no idea of the consequences.

Yet a heavily virtualized envirnment makes such events inevitable. It takes tools like TeamQuest to provide enough visibility to detect and prevent similar occurrences.

Craig

The afternoon sessions at TTS were very informative.  Rey Rios gave a great session on ITSO and ITIL.  He discussed the differences, and similarities, between ITIL V2, V3 and ITSO.  Then he explained how TeamQuest products supported the frameworks and best practices.

Mark Manness and Scott Johnson followed Rey.  They discussed the upcoming changes to the VMWare agent and TeamQuest Model as it relates to modeling VMWare platforms and guests.  It was very informative and all are awaiting the releases.

Leonid Grinshpan of Oracle discussed multi-tiered application sizing.  It was great presentation.  Not only did he give us insight into modeling Oracle, he discussed step-by-step processes on how to build prototyping models without and system data, how to validate them and apply what-if scenarios.  I thought he did a great job of showing us how we can prototype new applications before they are built.

TTS seems to get better every year and this year is no exception.

Until the next time…

Ron

Jacques Lehoullier, senior architect for product selcetion at ING Canada (soon to be known as Intact Financial Corporation) gave a breakout session on how to expand AIX capacity from a fixed physical base. He laid out the fact that rapid growth led to increased hardware costs, higher logistical costs and delays due to floor space, electrical and cooling issues. The data center had literally ran out of space.

The company investigated virtualization as a solution, but worried about the risks due to having many applications running on one physical server. It upgraded from two IBM P595 servers of 64 CPUs each to four P590+ with 16 CPUs each in a virtual environment consisting of four CPU pools.

IT at the company consulted with Gartner Group which recommended TeamQuest. This helped the company simplify reporting — from as much as a month to gather the performance and capacity data down to seconds — and enabled the company to decommission several old servers despite adding 100 more logical partitions.  

Craig.