Ping to test your SAN
It isn't good enough nowadays to just add more hardware onto your system when you see something slowing down. Slow transaction rates or poor Web site performance might mean a problem in your servers, or it could mean that your SAN isn't operating as well as it could. How to find out? This tip suggests one way that you can see if it's your SAN that needs its oil changed, or if some other system is the culprit.
Got a storage admin tip of your own? Why not
Although most SAN tuning is done either at the time of setup or when pre-set thresholds are breached, indicating a problem, it's good insurance to test a SAN's performance on a regular basis.
One of the most useful tools for doing this is a feature in some SAN management software that works much like the ping command on a Unix system. That is, it tests the latency of the system by sending a small chunk of data (sometimes called a 'bullet') from an initiating host to a target device. By comparing actual to theoretical latency in the data transfer, as well as comparing the current performance with previous performance, the administrator can spot slowdowns before they become bottlenecks.
One common source of increased latency is changes in the configuration of the SAN. Like most storage systems, SANs change constantly as new devices are added and new demands are placed on the storage system. Often these changes will have subtle (and sometimes not-so-subtle) effects on the performance of the SAN.
Sanavigator (www.sanavigator.com) has a white paper on its Web site titled "Optimizing SAN Performance" which gives an overview of how to tune a SAN for best performance.
Did you like this tip? Why not let us know. Email to share your opinion.
Rick Cook has been writing about mass storage since the days when the term meant an 80K floppy disk. The computers he learned on used ferrite cores and magnetic drums. For the last twenty years he has been a freelance writer specializing in storage and other computer issues.
This was first published in October 2001