Title :
Accelerating I/O Forwarding in IBM Blue Gene/P Systems
Author :
Vishwanath, Venkatram ; Hereld, Mark ; Iskra, Kamil ; Kimpe, Dries ; Morozov, Vitali ; Papka, Michael E. ; Ross, Robert ; Yoshii, Kazutomo
Author_Institution :
Argonne Nat. Lab., Argonne, IL, USA
Abstract :
Current leadership-class machines suffer from a significant imbalance between their computational power and their I/O bandwidth. I/O forwarding is a paradigm that attempts to bridge the increasing performance and scalability gap between the compute and I/O components of leadership-class machines to meet the requirements of data-intensive applications by shipping I/O calls from compute nodes to dedicated I/O nodes. I/O forwarding is a critical component of the I/O subsystem of the IBM Blue Gene/P supercomputer currently deployed at several leadership computing facilities. In this paper, we evaluate the performance of the existing I/O forwarding mechanisms for BG/P and identify the performance bottlenecks in the current design. We augment the I/O forwarding with two approaches: I/O scheduling using a work-queue model and asynchronous data staging. We evaluate the efficacy of our approaches using microbenchmarks and application-level benchmarks on leadership class systems.
Keywords :
multiprocessing systems; parallel machines; scheduling; I/O bandwidth; I/O forwarding mechanism; I/O scheduling; I/O subsystem; IBM Blue Gene/P systems; application-level benchmark approach; asynchronous data staging; leadership class systems; leadership computing facilities; leadership-class machines; microbenchmark approach; work-queue model; Benchmark testing; Computer architecture; Instruction sets; Lead; Servers; Switches; Throughput;
Conference_Titel :
High Performance Computing, Networking, Storage and Analysis (SC), 2010 International Conference for
Conference_Location :
New Orleans, LA
Print_ISBN :
978-1-4244-7557-5
Electronic_ISBN :
978-1-4244-7558-2
