Production I / O Characterization on the Cray XE 6

I/O performance is an increasingly important factor in the productivity of large-scale HPC systems such as Hopper, a 153,216 core Cray XE6 system operated by the National Energy Research Scientific Computing Center. The scientific workload diversity of such systems presents a challenge for I/O performance tuning, however. Applications vary in terms of data volume, I/O strategy, and access method, making it difficult to consistently evaluate and enhance their I/O performance. We have adapted the Darshan I/O characterization tool for use on Hopper in order to address this challenge. Darshan is an I/O instrumentation library that collects I/O access pattern information from large-scale production applications with minimal overhead. In this paper we present our experiences in deploying Darshan on the Cray XE6 platform, including performance evaluation of Darshan with up to 98,304 processes and a case study of how to identify applications that can benefit most from I/O performance tuning. Darshan was automatically enabled for all Hopper users in November 2012 and instruments over 5,000 jobs per day as of April 2013.