Improved Parallel I/O via a Two-phase Run-time Access Strategy * (Presented in: IPPS ‘93 Parallel I/O Workshop)
نویسندگان
چکیده
In recent years, two important events have occurred in the area of high-performance computing: the development of very high speed processors, and the development of massively parallel computers based on these processors. The processing capacities of these parallel computers have made them the computational instrument of choice in the scientific community and they can be found in one form or another in almost every major academic and research institution. Some of the commercially available parallel computers include Intel Paragon [INTE92], nCUBE [NCUB92], CM-5 [THIN91], and Kendall Square [KSR92]. It is anticipated that within the next few years, parallel computers will be capable of computational rates in the Teraflops range [HPCC91]. Abstract
منابع مشابه
Reducing I/O Complexity by Simulating Coarse Grained Parallel Algorithms
Block-wise access to data is a central theme in the design of efficient external memory (EM) algorithms. A second important issue, when more than one disk is present, is fully parallel disk I/O. In this paper we present a deterministic simulation technique which transforms parallel algorithms into (parallel) external memory algorithms. Specifically, we present a deterministic simulation techniq...
متن کاملParallel Algorithms for Hypercube Allocation
Parallel algorithms of the hypercube allocation strategies are considered in this paper. Although the sequential algorithms of various hypercube allocation strategies are easier to implement, their worst case time complexities exponentially increase as the dimension of the hypercube increases. We show that the free processors can be utilized t o perform the allocation jobs in parallel to improv...
متن کاملMTIO - A Multi-Threaded Parallel I/O System
This paper presents the design and evaluation of a multithreaded runtime library for parallel I/O. We extend the multi-threading concept to separate the compute and I/O tasks in two separate threads of control. Multi-threading in our design permits a) asynchronous I/O even if the underlying file system does not support asynchronous I/O; b) copy avoidance from the I/O thread to the compute threa...
متن کاملData Access Reorganizations in Compiling Out-of-Core Data Parallel Programs on Distributed Memory Machines
This paper describes optimization techniques for translating out-of-core programs written in a data parallel language like HPF to message passing node programs with explicit parallel I/O. We rst discuss how an out-of-core program can be translated by extending the method used for translating in-core programs. We demonstrate that straightforward extension of in-core compilation techniques does n...
متن کاملTesting a Simple Polygon for Monotonicity Optimality in Parallel
Sumanta Guhat We show that, in parallel, an n-vertex simple polygon can be tested for monotonicity optimally in O(logn) time using O(n/logn) EREW PRAM processors, and we present two different optimal parallel algorithms for solving this problem. OUf result leads to an optimal parallcl algorithm for triangulating simple polygons that runs in O(logn) time using O(n/logn) EREW PRA!\I processors if...
متن کامل