Influence of Inter-Agent Communication Cost to Diffusion Scheduling in Irregular Parallel Computations

نویسندگان

  • Marek Grochowski
  • Ewa Tusha
  • Piotr Uhruski
چکیده

The paper briefly presents the architecture of a multi-agent computing system with an emphasis on diffusion scheduling. Two binding energy formulas used by the diffusion scheduling algorithm are introduced: one respecting only requirements for the CPU and memory, and a second also utilizing the requirement for inter-agent communication. Two pairs of experiments with irregular parallel computations are presented with comparative results showing the efficiency in respect to inter-agent communication in both local and wide area networks.

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

Scheduling of Multiple Autonomous Guided Vehicles for an Assembly Line Using Minimum Cost Network Flow

This paper proposed a parallel automated assembly line system to produce multiple products having multiple autonomous guided vehicles (AGVs). Several assembly lines are configured to produce multiple products in which the technologies of machines are shared among the assembly lines when required. The transportation between the stations in an assembly line (intra assembly line) and among station...

متن کامل

MULTIPROCESSOR SCHEDULING TO ACCOUNT FOR INTERPROCESSOR COMMUNICATION by Gilbert

Interprocessor communication (PC) overheads have emerged as the major performance limitation in parallel processing systems, due to the transmission delays, synchronization overheads, and conflicts for shared communication resources created by data exchange. Accounting for these overheads is essential for attaining efficient hardware utilization. This thesis introduces two new compile-time heur...

متن کامل

Free Scheduling of General Nested Loops For Distributed Memory Architectures

The most extensive, in terms of time execution, part of a program is the nested loops. Loop parallelization involves two steps: First the time partitioning of the index space to achieve the minimum makespan, and second the efficient assignment of the concurrent partitions into the target parallel architecture. If distributed memory multiprocessor systems are used, overall performance is decline...

متن کامل

Improving the Performance of Large-Scale Unstructured PDE Applications

This paper investigates two types of overhead due to duplicated local computations, which are frequently encountered in the parallel software of overlapping domain decomposition methods. To remove the duplication-induced overhead, we propose a parallel scheme that disjointly re-distributes the overlapping mesh points among irregularly shaped subdomains. The essence is to replace the duplicated ...

متن کامل

An Overview of the RAPID Run-time System for Parallel Irregular Computations

RAPID is a run-time system that uses an inspector/executor approach to parallelize irregular computations by embodying graph scheduling techniques to optimize interleaved communication and computation with mixed granularities. It provides a set of library functions for specifying irregular data objects and tasks that access these objects, extracts a task dependence graph from data access patter...

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:
  • Inteligencia Artificial, Revista Iberoamericana de Inteligencia Artificial

دوره 9  شماره 

صفحات  -

تاریخ انتشار 2005