Modeling Communication Locality in Multiprocessors

نویسندگان

  • Charles A. Salisbury
  • Z. Chen
  • Rami G. Melhem
چکیده

Locality of reference is an important aspect of many computer operations. It is often exploited to optimize the performance of computer functions. In this paper, we apply the locality concept to the communication patterns of parallel programs operating over an interconnection network with a fixed communication latency between any pair of attached nodes. Unbuffered multistage networks and all-optical networks are examples of these. We quantify the notions of spatial and temporal locality in this context, and combine them in a locality measure. This measure is used as the basis for identifying the communication working sets of a parallel program. We focus on programs with a looping structure and investigate conditions under which each working set consists of the complete set of paths required by a single loop. 1999 Academic Press

برای دانلود رایگان 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

Bidirectional Ring: An Alternative to the Hierarchy of Unidirectional Rings

A hierarchy of unidirectional rings has been used successfully in distributed shared-memory multiprocessors. The xed cluster size of the hierarchy prevents full exploitation of communication locality. The bidirectional ring is presented as an alternative to the hierarchy. Its relative performance is evaluated for a variety of memory access patterns and network sizes. It gives superior performan...

متن کامل

Enhancing the Performance of Autoscheduling with Locality-Based Partitioning in Distributed Shared Memory Multiprocessors

Abstract. Autoscheduling is a parallel program compilation and execution model that combines uniquely three features: Automatic extraction of loop and functional parallelism at any level of granularity, dynamic scheduling of parallel tasks, and dynamic program adaptability on multiprogrammed shared memory multiprocessors. This paper presents a technique that enhances the performance of autosche...

متن کامل

Locality-based Scheduling for Shared-memory Multiprocessors Current and Future Trends in Parallel and Distributed Computing

The last decade has produced enormous improvements in microprocessor performance without a corresponding improvement in memory or intercon-nection network performance. As a result, the relative cost of communication in shared-memory multiprocessors has increased dramatically. Although many applications could ignore the cost of communication and still achieve good performance on the previous gen...

متن کامل

Locality-Based Scheduling for Shared-Memory Multiprocessors

The last decade has produced enormous improvements in microprocessor performance without a corresponding improvement in memory or interconnection network performance. As a result, the relative cost of communication in shared-memory multiprocessors has increased dramatically. Although many applications could ignore the cost of communication and still achieve good performance on the previous gene...

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

عنوان ژورنال:
  • J. Parallel Distrib. Comput.

دوره 56  شماره 

صفحات  -

تاریخ انتشار 1999