What are the challenges of implementing a high-performance NAS solution for a client, and how can solution providers overcome those challenges?
Providers don't necessarily match the high performance of a solution to the client's requirements. Solution providers need to take the time to listen to their clients, understand their data sets and then match the two properly. The challenge is that while many vendors claim to have "high-performance" NAS products, each offering has limitations. A solution provider doesn't want to let its client run up against those limitations after the sale -- it's better to identify those limitations before the sale.
 |
| About the author |
| Jason Sparks is VP of storage and systems at Xiologix. Listen to more from Jason in our high-performance NAS podcast |
|
|
|
|
In many cases, high-performance NAS solutions are based on NFS operations or IOPS or throughput, and each of those has different attributes that affect the client. For example, a client may need to support a large file system with millions of files per directory structure. In those cases, you're trying to offset metadata scanning within the file system, and that's why you move to a high-performance NAS system. But that has nothing to do with NFS operations or IOPS or even theoretical throughput or bandwidth. In other cases, clients in streaming media markets or high-performance computing (e.g., HPTC or HPEC) really do care about theoretical performance and throughput, and they opt for high-performance NAS that meets those particular requirements -- they don't necessarily care how many files can be sorted in a directory. Understand the limiting factors of a high-performance device to ensure that they do not impair the client in their environment.
Return to the high-performance NAS FAQ guide and read the rest of Jason's expert responses.
