Optimizing parallel I/O performance of HPC applications.

摘要

Parallel I/O is an essential component of modern High Performance Computing (HPC). Obtaining good I/O performance for a broad range of applications on diverse HPC platforms is a major challenge, in part because of complex inter-dependencies between I/O middleware and hardware. The parallel file system and I/O middleware layers all offer optimization parameters that can, in theory, result in better I/O performance. Unfortunately, the right combination of parameters is highly dependent on the application, HPC platform, and problem size/concurrency. Scientific application developers do not have the time or expertise to take on the substantial burden of identifying good parameters for each problem configuration. They resort to using system defaults, a choice that frequently results in poor I/O performance. We expect this problem to be compounded on exascale class machines, which will likely have a deeper software stack with hierarchically arranged hardware resources.;We present a line of solution to this problem containing an autotuning system for optimizing I/O performance, I/O performance modeling, I/O tuning, I/O kernel generation, and I/O patterns. We demonstrate the value of these solution across platforms, applications, and at scale.