Getting started with I/O profiling

Profiles in Storage

Article from ADMIN 08/2012

By Jeff Layton

The selection of various HPC storage solutions is one of the most pressing issues in HPC, requiring research, study, and planning to be effective, which usually means understanding how your applications perform I/O. We present some techniques for examining I/O patterns in HPC applications.

Storage is one of the largest issues – if not the largest issue – in high-performance computing (HPC). All aspects of HPC storage are critical to the overall success or productivity of systems: high performance, high reliability, access protocols, scalability, ease of management, price, power, and so on. These aspects, and, perhaps more importantly, combinations of these aspects, are key drivers in HPC systems and performance.

With so many options and so many key aspects to HPC storage, a logical question you might ask is: Where should I start? Will a NAS (Network Attached Storage) solution work for my system? Do I need a high-performance parallel filesystem? Should I use a SAN (Storage Area Network) as a back end for my storage, or can I use less expensive DAS (Direct Attached Storage)? Should I use InfiniBand for the compute node storage traffic or will GigE or 10GigE be sufficient? Should I use 15,000rpm drives or 7,200rpm drives? Do I need SSDs (solid state drives)? Which filesystem should I use? Which I/O scheduler should I use within Linux? Should I tune my network interfaces? How can I take snapshots of the storage, and do I really need snapshots? How can I tell if my storage is performing well enough? How do I manage my storage? How do I monitor my storage? Do I need a backup or just a copy of the data? How can I monitor the state of my storage? Do I need quotas, and how do I enforce them? How can I scale my storage in terms of performance and capacity? Do I need a single namespace? How can I do a filesystem check, how long will it take, and do I need really one? Do I need cold spare drives or storage chassis? What RAID level do I need (file or object)? How many hot spares are appropriate? SATA versus SAS? And on and on.

When designing or selecting HPC storage, these are some of many questions to consider, but you might notice one item I left out of this laundry list: I did not discuss applications.

Designing HPC storage, just as

...

Use Express-Checkout link below to read the full article (PDF).