Has the following scenario ever occurred in your IT shop?
Joe the systems administrator sticks his head into your office and says, "FYI, we are at 97% capacity on our Exchange Server and I'm getting errors from the box. "
If so, your capacity planning efforts have failed. At this point there isn't much you can do. You are in a no-win situation. More than likely, you will have to go ask for money that wasn't budgeted, and in these economic times, you never know how damaging this will be.
While you can't turn back the hands of time, you can change your future. Here is a recommended methodology for developing a capacity planning strategy to help you get out of a nightmare and avoid the next one:
- Phase I: Solve the right problem at the right time
- Phase II: Short-term execution
- Phase III: Long-term planning
Phase I: Solve the right problem at the right time
All too often we view the world in tunnel vision. We are presented with a problem we solve that specific problem and then move on.
Unfortunately, addressing a situation with tunnel vision typically alleviates one bottleneck only to expose another. Most capacity management problems are complex and have several answers. In some cases, capacity management is not the problem at all. This I learned firsthand…
A few years ago, a business owner asked me to run a monthly database performance report. Over the course of a few months, he had seen performance degradation, prompting him to raise concerns and become very vocal about the situation. His knee-jerk reaction was to express concern about database performance.
The fire drill began and every fix, from replacing the server to upgrading memory, was considered. As it turned out, the problem was not that the database was slow, but that the report wasn't frequent enough, which caused minor performance spikes to unfairly impact the overall report. We solved our "capacity" issue by running the report more frequently.
Bottom line: Take a step back from the problem to identify all the facts and then work toward the right solution.
Phase II: Short-term execution
After you've established the cause, you can begin to plan for immediate relief. In the Exchange crisis I mentioned in the beginning of this article, we ended up using the Microsoft EXBPA tool to analyze Exchange Server and quickly identified several users with extremely large inboxes.
Bottom line: Look at the facts to see where the short-term wins are and how they can help get you out of crisis mode.
Phase III: Long-term planning
Too often we leave the short-term execution phase and go back to the daily grind and quickly forget that we didn't really solve the problem. Sure, we brought the Exchange utilization rate to 85%, down from 97% capacity, but it was a short-term fix.
In this instance, we needed long-term strategies, i.e. mailbox size restrictions, archiving ex-employee data, a scheduled upgrade to the email server (on our own terms), and a system to report capacity metrics to avoid ever getting to 90% utilization.
Bottom line: Unless long-term strategies are put into place, you'll end right back where you started.
Also note, aside from these strategies there are many tools that can help you in your capacity–planning efforts -- the exact tool and methodology will depend on your situation. Microsoft offers several tools, including EXBPA and Microsoft System Center. The same tools used to get you out of a bind can often help you stay out of a crisis.
ABOUT THE AUTHOR
Russell Olsen is an IT professional with a solid business foundation. He has a wide range of experience, including CIO, VP of product development, VP of operations and senior auditor for a Big Four accounting firm performing technology risk assessments and Sarbanes-Oxley audits. Russell is a CISA, GSNA and MCP.
This was first published in March 2009