Improving Performance by Use of Adaptive Objects: Experimentation with a Configurable Multiprocessor Thread Package
نویسندگان
چکیده
Operating system kernels typically ooer a xed set of mechanisms and primitives. However, recent research shows that the attainment of high performance for a variety of parallel applications may require the availability of variants of existing primitives or additional low-level mechanisms best suited for speciic applications. One approach to addressing this need is to ooer an adaptable and extensible operating system kernel. In this paper, we present a model for adaptive objects and associated mechanisms which may be used for the development of high performance operating system kernels for parallel and distributed systems. We use the model to implement a class of multiprocessor locks namely, adaptive locks which adapt themselves according to user-provided adaptation policies to suit any application locking pattern. Using a well-known multiprocessor application, the Travelling Sales Person program, we demonstrate the performance advantage of adaptive locks over existing locks.
منابع مشابه
Experimentation with Configurable, Lightweight Threads on a KSR Multiprocessor
The implementation of operating system functions can significantly affect the performance of parallel programs. In this paper, we describe our experience with a reconfigurable, multiprocessor threads package on a 32-node KSR1 supercomputer. We discuss static and dynamic configurations that address the on-line reconfiguration of threads schedulers and synchronization constructs. Experimental res...
متن کاملA Portable Multiprocessor Interface for Standard Ml of New Jersey
We have designed a portable interface between shared-memory multiprocessors and Standard ML of New Jersey. The interface is based on the conventional kernel thread model and provides facilities that can be used to implement user-level thread packages. The interface supports experimentation with diierent thread scheduling policies and synchronization constructs. It has been ported to three diier...
متن کاملImplementation of Scalable Blocking Locks Using an Adaptive Thread Scheduler
Blocking locks are commonly used in parallel programs to improve application performance and system throughput. However, most implementations of such locks suuer from two major problems { latency and scalability. In this paper, we propose an implementation of blocking locks using scheduler adaptation which exploits the interaction between thread schedulers and locks. By experimentation using we...
متن کاملToward an automatic mapping of DSP algorithms onto parallel processors
With ever increasing computational requirements of complex DSP algorithms and applications, implementation on multiprocessor platforms becomes a necessity. The main problem is lack of necessary software tools for multiprocessor mapping. We present the main features of a prototype design environment which allows direct mapping of complex DSP applications, designed for implementation on a single ...
متن کاملWSSMP: A High-Performance Shared- and Distributed-Memory Parallel Sparse Symmetric Linear Equation Solver
The Watson Symmetric Sparse Matrix Package, WSSMP, is a high-performance, robust, and easy to use software package for solving large sparse symmetric systems of linear equations. It can be used as a serial package, or in a shared-memory multiprocessor environment, or as a scalable parallel solver in a message-passing environment, where each node can either be a uniprocessor or a shared-memory m...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1993