A Case Study in Using Local I/O and GPFS to Improve Simulation Scalability
نویسندگان
چکیده
Many optimization algorithms exploit parallelism by calling multiple independent instances of the function to be minimized, and these function in turn may call off-the-shelf simulators. The I/O load from the simulators can cause problems for an NFS file system. In this paper we explore efficient parallelization in a parallel program for which each processor makes serial calls to a MODFLOW simulator. Each MODFLOW simulation reads input files and produces output files. The application is "embarassingly" parallel except for disk I/O. Substitution of local scratch as opposed to global file storage ameliorates synchronization and contention issues. An easier solution was to use the high performance global file system GPFS instead of NFS. Compared to using local I/O, using a high performance shared file system such as GPFS requires less user effort.
منابع مشابه
Using IKAROS to provide Scalable I/O bandwidth
We present IKAROS as a utility that permit us to form scalable storage platforms. IKAROS enable us to create ad-hoc nearby storage formations and use a huge number of I/O nodes in order to increase the available bandwidth. We measure the performance and scalability of IKAROS versus the IBM’s General Parallel File System (GPFS) under a variety of conditions. The measurements are based on benchma...
متن کاملStorm, an Srm Implementation for Lhc Analysis Farms
In production grids, high performance disk storage solutions using parallel file systems are becoming increasingly important to provide reliability and high speed I/O operations needed by HEP analysis farms. Today, Storage Area Network solutions are commonly deployed at LHC center and and parallel file systems such as GPFS and Lustre allow for reliable, high-speed native POSIX I/O operations in...
متن کاملA Case Study: Performance Analysis and Optimization of SAS® Grid Computing Scaling on a Shared Storage
SAS® Grid Computing is a scale-out SAS® solution that enables SAS applications to better utilize I/O and compute intensive computing resources. This requires the use of high-performance shared storage (SS) that allows all servers to access the same file system. SS may be implemented via traditional NFS NAS or clustered file systems (CFS) like GPFS. This paper uses the Intel® Enterprise Edition ...
متن کاملPatterns of flow evolution in the central area of the Romanian Plain, Case study: the Calnistea Catchment (Romania)
This paper seeks to emphasize the flow variability in the Calnistea catchment by analyzing the local physiographic factors. The research has shown that the amount of precipitation that falls to the ground is low, the rocks in the region are soft, but highly permeable, gradients are gentle in most of the territory and vegetal cover is sparse and therefore cannot hold important amounts of water. ...
متن کاملIntroduction to Linux Cluster Filesystems
Traditional local file systems support a persistent name space. A local file system views devices as being locally attached, the devices are not shared, and hence there is no need in the file system design to enforce devicesharing semantics. Instead, the focus is on aggressively caching and aggregating file system operations to improve performance by economizing on the number of actual disk acc...
متن کامل