A nity - based Self Scheduling for Software Shared Memory System 1

نویسندگان

  • Weisong Shi
  • Zhimin Tang
  • Weiwu Hu
چکیده

In this paper, we propose an aanity-based self scheduling scheme (ABS) for software shared memory system. In this scheme, the static aanity between processor and initial data distribution is considered when scheduling, and the synchronization overhead is reduced greatly when load imbalance occurs. Comparized with previous schemes, ABS performs better in metacomputing environment.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Issues in Implementation of Cache - A nity Scheduling Murthy

In a shared memory multiprocessor, a thread may have an a nity to a processor because of the data remaining in the processor's cache from a previous dispatch. We show that two basic problems should be addressed in a Unix-like system to exploit cache a nity for improved performance: First, the limitation of the Unix dispatcher model (\processor seeking a thread"); Second, pseudo-a nity caused by...

متن کامل

A More Practical Loop Scheduling Scheme for Home-based Software DSMs

Loop scheduling is a critical issue for achieving good performance in parallel and distributed systems. However, this issue is neglected in the area of software DSM systems in the past. All the previous research is based on dedicated homogeneous hardware environment, such as shared memory multiprocessors, and hardware distributed shared memory systems. The main contribution of this paper is two...

متن کامل

The E ectiveness of A nity - Based Scheduling inMultiprocessor Networking ( Extended

| Techniques for avoiding the high memory overheads found on many modern shared-memory multiproces-sors are of increasing importance in the development of high-performance multiprocessor protocol implementations. One such technique is processor-cache aanity scheduling, which can signiicantly lower packet latency and substantially increase protocol processing throughput 30]. In this paper, we ev...

متن کامل

Performance Optimization of Software Distributed Shared Memory Systems

Software Distributed Shared Memory Systems (DSMs, or Shared Virtual Memory) are advocated to be an ideal vehicle for parallel programming because of its combination of programmability of shared memory and scalability of distributed memory systems. The challenge in building a software DSM system is to achieve good performance over a wide range of parallel programs without requiring programmers t...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 1999