High Availability Storage Options and Their Impact on Performance

9
"Rules of Thumb"
for Improving MPE/iX Disk Performance
What I would like to do now is list some guide lines to help make the migration to an array a better
experience. These are not hard rules and some flexibility is needed as you incorporate these "Rules of
Thumb" into your environment. The most important thing is to help set realistic expectations as storage
technology changes and improves:
1. MPE/iX prefers many small Ldevs over a few large Ldevs. Configure your array luns in the
preferred 4Gbytes to 18Gbytes range. Also create (more) Ldevs for the system volume set. This
allows MPE/iX to "spread" the file extents and system transient space (the swap area) across a
number of Ldevs (MPE's extent striping implementation). This will have an improved performance
over a configuration which uses a single large Ldev.
2. Restore the application’s data to User Volumes. Create as many User Volume Sets as makes sense.
This reduces the XM bottleneck to that of the master volume of the volume set, adds more pathways
to the data and adds fault containment. This also reduces the amount of data to reload in case of
catastrophic disk failure.
3. Gather data on the user’s application I/O characteristics. There are several tools available like
GLANCE/XL that can help collect this data.
4. Do not configure more that 16 Ldevs per fibre channel pathway or bus and make sure the number of
Ldevs per pathway or bus is supported in that environment. (HVD-SCSI should be limited to 8 while
SE-SCSI should be no more than 4) Consolidating a large number of physical devices to a fewer
number of larger capacity devices might create a situation where the array is the bottleneck while
too many Ldevs per pathway or bus can cause the connection technology (FC) to be the bottleneck.
(Your mileage will vary depending on driving habits)
5. Purchase a high availability disk drive technology for its feature set and its ability to protect data.
Performance of these products is dependent on a clear understanding of how it is used, I/O
characteristics of the application and the way it is configured.
---------------------------