Exploration of Parallel Storage Architectures for a Blue Gene/L on the TeraGrid
نویسندگان
چکیده
This paper examines the construction of a cost-effective storage cluster for an IBM Blue Gene/L supercomputer on the TeraGrid. The performance of storage clusters constructed using commodity Opteron servers and directly-attached fibre-channel RAID systems is analyzed utilizing the GPFS, Lustre, and PVFS file systems. In addition to traditional client scalability results, server scalability is also examined, identifying the number of servers required to saturate a 10 Gbps infrastructure network uplink. The results highlight test and system configurations that provide scalable throughput and identify configurations where the environment restricts scalability. Moreover, the Blue Gene/L’s multiplexed I/O design introduces an additional point of contention that affects each file system to varying degrees.
منابع مشابه
Influences of Device Architectures on Characteristics of Organic Light-Emitting Devices Incorporating Ambipolar Blue-Emitting Ter(9,9-diarylfluorenes)
In this article, we report the studies of various device architectures of organic lightemitting devices (OLEDs) incorporating highly efficient blue-emitting and ambipolar carriertransport ter(9,9-diarylfluorene)s, and their influences on device characteristics. The device structures investigated include single-layer devices and multilayer heterostructure devices employing the terfluorene as one...
متن کاملModel and simulation of exascale communication networks
Exascale supercomputers will have millions or even hundreds of millions of processing cores and the potential for nearly billion-way parallelism. Exascale compute and data storage architectures will be critically dependent on the interconnection network. The most popular interconnection network for current and future supercomputer systems is the torus (e.g., k-ary, n-cube). This paper focuses o...
متن کاملEfficient High-precision Dense Matrix Algebra on Parallel Architectures for Nonlinear Discrete Optimization
We provide a proof point for the idea that matrix-based algorithms for discrete optimization problems, mainly conceived for proving theoretical efficiency, can be easily and efficiently implemented on massively-parallel architectures by exploiting scalable and efficient parallel implementations of algorithms for ultra high-precision dense linear algebra. We have successfully implemented our alg...
متن کاملInterweaving Data and Computation for End-to-End Envi- ronmental Exploration on the TeraGrid
This paper presents the design and implementation of a cyberinfrastructure for End-to-End Environmental Exploration (C4E4). The C4E4 framework addresses the need for an integrated data/computation platform for studying broad environmental impacts by combining heterogeneous data resources with state-of-the-art modeling and visualization tools. With Purdue being a TeraGrid Resource Provider, C4E4...
متن کاملBenchmarking Parallel I/O Performance for a Large Scale Scientific Application on the TeraGrid
This paper is a report on experiences in benchmarking I/O performance on leading computational facilities on the NSF TeraGrid network with a large scale scientific application. Instead of focusing only on the raw file I/O bandwidth provided by different machine architectures, the I/O performance and scalability of the computational tools and libraries that are used in current production simulat...
متن کامل