Sorting on Clusters of SMPs

نویسندگان

  • David R. Helman
  • Joseph JáJá
چکیده

We introduce an efficient algorithm for sorting on clusters of symmetric multiprocessors (SMPs). This algorithm relies on a novel scheme for stably sorting on a single SMP coupled with balanced regular communication on the cluster. Our SMP algorithm seems to be asymptotically faster than any of the published algorithms. The algorithms were implemented in C using POSIX threads and the SIMPLE library of communication primitives and run on a cluster of DEC AlphaServer 2100A systems. Our experimental results verify the scalability and efficiency of our proposed solution and illustrate the importance of considering both memory hierarchy and the overhead of shifting to multiple nodes.

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

SIMPLE: A Methodology for Programming High Performance Algorithms on Clusters of Symmetric Multiprocessors (SMPs)

We describe a methodology for developing high performance programs running on clusters of SMP nodes. The SMP cluster programming methodology is based on a small prototype kernel (SIMPLE) of collective communication primitives that make efficient use of the hybrid shared and message passing environment. We illustrate the power of our methodology by presenting experimental results for sorting int...

متن کامل

An Introduction to Balder - An OpenMP Run-time Library for Clusters of SMPs

In this paper a run-time library, called Balder, for OpenMP 2.0 is presented. OpenMP 2.0 is an industry standard for programming shared memory machines. The run-time library presented can be used on SMPs and clusters of SMPs and it will provide a shared address space on a cluster. The functionality and design of the library is discussed as well as some features that are being worked on. The per...

متن کامل

Modeling Cone-Beam Tomographic Reconstruction Using LogSMP: An Extenced LogP Model for Clusters of SMPs

The tomographic reconstruction for cone-beam geometries is a computationally intensive task requiring large memory and computational power to investigate interesting objects. The analysis of its parallel implementation on widely available clusters of SMPs requires an extension of the original LogP model to account for the various communication channels, called LogSMP. The LogSMP model is used i...

متن کامل

Technische Universität Chemnitz Sonderforschungsbereich 393 Numerische Simulation auf massiv parallelen Rechnern

The characteristics of irregular algorithms make a parallel implementation difficult, especially for PC clusters or clusters of SMPs. These characteristics may include an unpredictable access behavior to dynamically changing data structures or strong irregular coupling of computations. Problems are an unknown load distribution and expensive irregular communication patterns for data accesses and...

متن کامل

Home-based SVM Protocols for SMP Clusters: Design, Implementation and Performance

Shared virtual memory has been used as a cost-eeective means of providing the attractive shared-memory model to programmers. In the past, most SVM implementations have been focussed on clusters of uniprocessor systems. Due to the increasing popularity and declining costs of small scale shared memory multiprocessor systems (SMPs), they have become attractive candidates for replacing the uniproce...

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:
  • Informatica (Slovenia)

دوره 23  شماره 

صفحات  -

تاریخ انتشار 1998