When I read the eWeek article entitled, “How to Manage Capacity in Virtualized Environments” I finished wanting more. Actually, I finished realizing TeamQuest already has more to offer. 

Explaining the ITIL definition of Capacity Management is great, but how about expanding a bit further and drilling down into the details – from a Capacity Management perspective. Since the author, Rob Smoot, is a Group Product Marketing Manager at VMware, let’s focus on VMware.

TeamQuest offers scalable, agentless performance data collection for both VMware ESX and ESXi hypervisors and:

• vSphere 4
• AIX PowerVM, including LPARs and WPARs
• Solaris Containers, Zones, and LDOMs. 

TeamQuest Model, one tool in our suite, can predict performance in advance, help to optimize services and avoid costly bottlenecks altogether. We can find the least-expensive, best-performing configurations for P2V consolidation, ensuring that service levels will be met. IT managers can quickly determine the best way to allocate resources to VMware guests. And unlike other tools, TeamQuest Model understands response times inside virtual machines.

Take a look at a previous blog post entitled, “It’s CRAZY to Run Just One App Per Virtual Server.” Ron Potter, TeamQuest’s Manager of Best Practices outlines the benefits of running multiple applications on a virtual machine and the planning needed to ensure you have the right configuration for the right situation. In fact, Potter takes this a step further in his white paper, “Grab More Post-virtualization Savings.”

Also, check out the press release on TeamQuest’s VMware capabilities, or our website on managing a virtual environment, or for some real nitty gritty details, see our white paper entitled, “Managing a Virtual Computing Environment – How TeamQuest Supports VMware’s Virtual Machines.”

Remember virtualization is not a cure-all. Capacity Management improves your virtualization tools. It’s a discipline that, when used with the right tools, allows you to get the most from your virtualization efforts.

If you would like to further the discussion on how to truly manage capacity in virtualized environments, don’t hesitate to contact us directly or leave a comment below. We look forward to hearing from you!

Many organizations operate IT in firefighting mode, reacting to performance incidents when they occur. It is surprisingly less often that organizations perform capacity planning up front, though that would ensure that systems will provide necessary service levels without interruption. Being reactive is more expensive, but nonetheless, more common than being proactive. Maybe it’s human nature. It seems that few organizations possess the maturity to see the value of formally addressing capacity issues in an organized manner.

Maybe that’s changing.

A recent Network World article, “10 best IT jobs right now,” predicts that “Capacity Manager” will be an up and coming job title in 2010. In the article Denise Dubie suggests that tough economic conditions combined with added complexity from virtualization should help influence companies to pay more attention to capacity planning. Evelyn Hubbert of Forrester and Cameron Haight of Gartner are both quoted as emphasizing the strategic importance of capacity management for organizations seeking to maximize the value provided to a business by its IT organization.

At TeamQuest we’ve got tools for both proactive and reactive performance management, but we always try to influence customers to take a more proactive, service point of view. We suggest that the organizations with the highest level of capacity management maturity analyze IT capacity in terms of the business value that it can bring to an organization. (See the recent post here on the TeamQuest Capacity Management Maturity Model).

Demands for more services are increasing and high performance and reliability are expected, yet budgets, labor and floor space are all being constrained. I guess nobody said running a data center was easy.

The need for increased efficiency and reliability points to capacity management processes. And like anything else, you have to crawl before you walk, but the value of capacity management practices is there. So what’s it take to adopt more mature processes?

TeamQuest has released our Capacity Management Maturity Model that identifies different levels of IT management competency.

 

Following the lead of Gartner’s IT Management Process Maturity Model, we’ve drilled down even further to highlight the levels of Capacity Management maturity specifically.

After all, moving to mature Capacity Management tools and processes lowers costs, improves service quality and increases IT productivity since staff can focus on the most important duties rather than fight fires.

Take a look at our latest white paper entitled, Introducing a Capacity Management Maturity Model. Let us know what you think! Post your comments below.

This morning at CMG in Grapevine, Texas, TeamQuest’s Rey Rios gave an excellent talk about TeamQuest IT Service Optimization (ITSO) and how it relates to the broader, more encompassing ITIL framework.

ITSO is really a focused subset of ITIL, making it easier to get quick wins than if you dive headlong into a full ITIL implementation. It is analogous to the way that TeamQuest is able to excel at providing performance tools. We take careful aim on issues that affect the efficiency of IT services, and try not to be distracted by other aspects of systems management. Integration points in TeamQuest software allow customers to use TeamQuest’s specialized tools with whatever service desk or CMDB solution they choose, for example.

ITSO provides a formula for implementing effective ITIL Capacity Management, and TeamQuest tools can aid in that process. This will enable you to balance cost with performance, using business priorities to provide the proper perspective.

For more information about ITSO and ITIL, check out the links in the blog entry about Rey’s earlier CMG presentation. There is a video version of Rey’s presentation available on the TeamQuest website. You can find it on the TeamQuest VideoSeries web page.

Pop quiz hot shot!

What type of data is needed to build capacity plans? Better yet, what is the difference between trending and modeling? And while you’re thinking about it, can you explain Little’s Law?

Answer these questions correctly and you might have what it takes to move your IT team toward a more mature IT Operations environment.

If you’re looking for answers, visit our website for more information on Capacity Management. If  you know these answers, take the 16-question quiz and test your Capacity Management knowledge.  

Post your results and tell us what you think of the Capacity Management quiz. 

Anybody with a dartboard can claim to have a server capacity planning tool. Unfortunately, companies selling dartboards for capacity planning aren’t likely to be very honest about the sophistication of their tools. It’s caveat emptor, let the buyer beware.

So how do you avoid buying a dartboard when accurate predictions is what you really need? What makes a good capacity planning tool? What should you look for in a capacity planning tool?

First off, monitoring performance is not capacity planning. Getting an alarm event 15 minutes before users complain does not constitute capacity planning. Monitoring and alarming are essential components for capacity management, but that’s not capacity planning.

Real time performance data isn’t enough either. Capacity planning always requires a historical record of some sort. But just keeping some historical data available for graphing and charting is not enough. You need more than that for accurate capacity planning.

Beware of companies touting trending as capacity planning. Computer system performance is not linear, and a capacity planning tool needs to know more than just past system performance in order to make accurate predictions about the future. And trending is of no use at all for many projects that involve capacity planning. You can’t plan a server consolidation project with trending, for example.

You should also be wary of tools that do “capacity planning” for server consolidation by adding together the resource utilization of each of the workloads that are being considered for possible consolidation. After normalizing CPU utilization to account for differences in computing capability, the utilization for each workload is added together to determine how much of the target CPU will be utilized after consolidation. A similar calculation is performed for other resources such as memory, I/O, and the network.

This kind of simplistic procedure can be good enough to find potential consolidation candidates, but it leaves way too much out of the equation for making the final decision when consolidating important workloads. You need a tool that understands something of the details regarding your server architecture, more about your applications’ use of that architecture, and how workloads will interact when they are consolidated.

OK, so I’ve said a lot about what isn’t a good capacity planning tool. You’re probably wondering what I would actually recommend. Not to say that less sophisticated methods don’t have their uses, but where critical apps are involved, you want a tool that performs capacity planning using some sort of modeling.

Sometimes when people talk about a “model” they mean a description or diagram. That”s not the kind of model I am talking about in this case. For sure, you need a description of the systems involved, but that description is really just a step in a good capacity planning process. What you want is a tool that can look at that description along with information regarding the incoming workloads, and predict how the systems will perform.

There are at least two types of methods used by capacity planning tools that use modeling to predict performance: simulation modeling and analytic modeling. A good simulation modeling tool will create a queuing network based on the system being modeled and pretend to run the incoming workloads on that network. Simulations like these can be very accurate, but a lot of work is necessary to adequately describe the systems with enough detail for the results to be dependable.

Analytic modeling also takes queuing into account, without pretending to run the incoming workloads on the model. Instead, in a good analytic modeling tool, formulas based on queuing theory are used to mathematically calculate processing times and delays. This type of modeling is much faster and not nearly so tedious to set up. And the results can be just as accurate as with simulation modeling.

Analytic models are not as generalized as what’s possible with simulation modeling, so when a crucial situation arises where a suitable analytic model is not available, it makes sense to put together a simulation model instead. The rest of the time you will want to stick with a much easier and faster analytic modeling process.

So analytic modeling is usually what you want in a capacity planning tool. If you really want to cover your bases, get a tool that can do both analytic and simulation modeling. And check that your capacity planning tool vendor isn’t mis-using the term “analytic” when making claims for their tool. You want to be sure that the tool you pick uses sound methods based on queuing theory to make its calculations, not something more closely resembling the less accurate capacity planning techniques I described in this article.

For more infomation, see teamquest.com/capacityplanning.

In his CMG 2008 Sunday workshop, “How High Will It Fly“, Dr Neil Gunther showed how relatively simple mathematical models fed with appropriate measurement data can be used to predict the scalability of a computing system. Unlike physical systems such as airplanes and bridges, computing systems don’t lose their wings or break apart when the load on the system exceeds the material strength.

Instead, a computing system’s performance starts to degrade; workload throughput levels off (less work is completed) and workload response times increase to infinity (work takes longer to complete, or may never complete!). 

One of the important and basic tasks for a performance analyst and capacity planner is to determine these critical limits – to understand the capability of the system so that it can be fully exploited without adverse effects. This is not a trivial task. And to make matters worse, today’s popular computing systems such as UNIX and Windows servers are multiprocessing systems.

In a single processing system the capability of that single processing unit directly affects the throughput and response time of the workload running on the system. A faster or more capable processing unit will yield improved results.

In contrast, in a multiprocessing system, to utilize the full capability of the system, the workload must be divided and coordinated between the multiple processing units.

For instance, several users may be updating the same table in a database. Although to the user, the update occurs immediately, the system coordinates the work so that only one user at a time is allowed to update the data. The other users have to wait their turn. This coordination work is plain and simple overhead; it’s time spent arranging work instead of completing work. This fact is a major reason why it is difficult to determine the critical limits of multiprocessing systems. Adding additional and faster processing units may not necessarily yield better results.

Gene Amdahl, one of the pioneers in this area, actually advocated for the use of single processing unit systems even though his famous law is most often quoted in papers on multiprocessing systems, according to Dr. Gunther. Perhaps he saw how much work lay ahead!

It is interesting that IBM seems to be producing systems with faster and faster processing units, whereas Sun Microsystems is producing systems with massively multithreaded processing units. Apparently two different strategies at work out in the market.

So determining how high it will fly is not trivial.

But there is hope! Software vendors such as TeamQuest, offer products to help performance analysts and capacity planners explore the limits of their increasingly more complex and powerful computing systems. TeamQuest Model was recently updated to fully understand the behavior of multiprocessing systems (CPUs, Cores per CPU, and Threads per Core.) And where simple models only give you the boundaries of the limits of the system, TeamQuest Model also provides the components of response that contribute to the overall response time of a workload.

Pascal

It was a very informative afternoon beginning with a discussion on the transition from Event Management to Business Service Management by Debra Curtis. This was followed up with a cloud computing presentation. But the most interesting discussion - in my opinion - was from Milind Govekar as he talked about performance management and capacity planning in a virtualized environment. 

Milind’s discussion was a very good presentation covering all aspects of Capacity Management – Monitoring, Historical Analysis, Capacity Planning and Tuning. He presented the business drivers for the work and explained the need to mature from Component Capacity planning to Service and ultimately Business Capacity Planning. It was reassuring to see TeamQuest was considered one of the major vendors providing the tools needed to support the space. He discussed the different aspects of the capacity planning role and urged attendees to make the role a priority as a large percentage of attendees had no formal capacity planning role. Listening to him, I heard everything we have been saying in our ITSO whitepapers, webinars, podcasts and white board series. It’s clear that they see things the same as we do!

Debra Curtis discussed how Infrastructure and Operations management needs to mature, keeping the component level view yet consolidating it into IT service views then on to Business service views. She discussed the current state of vendors and some new players and showed how other ITIL frameworks such as performance management feed information to the Event Correlation Analysis (ECA) and BSM tools.

She predicts that these new processes and tools will result in automatically determining what “normal” thresholds should be and be intelligent enough to alert only when operations and performance stray from those norms. Several times she stressed that tools alone cannot make you successful. Processes are important as well. I felt she reaffirmed our commitment to offer tools and our ITSO processes together to make our customers successful.

I listened to IBM discuss their current state in offering cloud computing. It is clear that they have thought this through and the complexities are far more than normal outsourcing arrangements. It appeared that they are very aggressive in this space. Only time will tell if cloud computing really will be the future of IT. 

More to come tomorrow 

Ron

The second day of the Gartner Data Center Conference has been informative. Besides the sessions, I have had some wonderful networking sessions. It appears that capacity planning/capacity management is becoming much more of a hot topic these days.

Not only is Gartner seeing more inquiries, many of the people I have spoken with in different IT operational areas are talking about the necessity of doing more formal capacity planning work. They all agree that the days of throwing hardware at capacity problems are behind us. Instead of over provisioning and continual firefighting efforts, we need to do a better job of planning for the future and spending more time tuning applications and services.

It is very satisfying to see that more and more people are beginning to see the value of the work we evangelize in TeamQuest IT Service Optimization.

Disruption in the Data Center

In the first keynote, Gartner analyst Carl Claunch discussed the “Top 10 Disruptive Technologies Affecting the Data Center.” They are in no particular order, according to Carl.

• Virtualization in storage

• Cloud computing

• Servers: Beyond the blade

• PC virtualization

• Enterprise Mash-ups

• Specialized systems

• Social Software and networking

• Unified communications

• Value of pods and zones

• Green IT

Additional Highlights from Claunch 

Carl gave us a lot of information, covering all the considerations, both pluses and minuses. I’ll cover a few items just to keep things brief. Gartner sees blade servers morphing into computing fabric where memory, I/O and processor are not restricted for use by a particular motherboard and can be redeployed at will. He states that the problem with blades is that when you outgrow one, that particular piece of technology is useless to you, becoming orphaned. You paid for the capacity and processing capabilities but can no longer take advantage of them. By taking technology to the next step by building computing fabrics, you have the ability to redeploy the component parts dynamically so that you continue to gain value from the asset.

Carl discussed enterprise mash-ups, pointing out how valuable they can be to organizations. A common deployment that we see frequently is linking an internet application to one of the mapping services to help your customers find their way to your place of business. Carl points out that this technology can be important in quickly building robust applications for use internally or externally. It can also help you attract and retain the new generation of digital workers. The challenge will be in managing the technology. Pieces of the application come from a variety or web sites which are outside your control so monitoring and measuring service performance becomes much more critical…and complex.

Sidenote
My last comment for this entry is about Don McMillan, an ex IEEE engineer and now a renown comedian. He gave the second keynote of the day. If you get the chance to see him, I highly recommend it. He had most of the crowd in tears. I can’t remember the last time I laughed so much. 

More to come. 

Ron

The benefits of server virtualization are great. We’ve heard most of it before - cost reduction, flexibility, etc. What IT professional wouldn’t be interested in consolidating under-utilized servers, reducing energy consumption or reducing the physical size of the data center?  Is there more we’re missing?

I believe Dan Kusnetzky brings up a good point: What’s next after server virtualization? We’re now getting into how do we understand and manage the virtual environment. I agree with Kusnetzky’s assessment, and I believe many organizations are missing one complementary discipline - Capacity Management. 

It’s the key to optimizing virtual environments. The constant flux and variables affecting services can affect IT resource allocation which makes Capacity Management important in virtual environments and to the business. How does IT know the impact of planned consolidation activities or understand what the application is doing inside the virtual machine?

In a step toward improving the management of virtual systems, we set out to findhow IT managers are utilizing virtualization in their organizations. It wasn’t surprising that respondents mentioned that proper planning is key for successful virtualization efforts. Forty percent of respondents noted that bottlenecks (events that impacted performance or end-user needs) are one of the top three challenges in deploying virtual systems. Fortunately, the survey also points out the effectiveness of tools to monitor and manage virtual systems as well as how organizations are measuring their efforts.  

You might be interested in more on the virtualization study. 

The Ovum survey results reinforce the point that up front Capacity Management is necessary to mitigate risk and maximize the ROI of virtualization efforts.  Take a look at this story from writer Denise Dubie called “7 tips for succeeding with virtualization.”  If you have any thoughts, success stories, or tips for those working on virtualization projects, feel free to post comments on the TeamQuest Blog.