An Evaluation of Parallel Computing in PC Clusters with Fast Ethernet

نویسندگان

  • Manuel E. Acacio
  • Óscar Cánovas Reverte
  • José M. García
  • Pedro E. López-de-Teruel
چکیده

Currently, clusters of Personal Computers (PCs) with sharing Fast Ethernet may o er a signi cant price/performance ratio in parallel processing. This paper presents a study of the promising possibilities of parallelism in this environment, using both a classical numerical method to solve a di erential equation and an innovative algorithm to solve the statistical problem of density estimation. In addition, and to compare, some comparisons with an IBM SP2 are carried out.

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

Cost-Performance Evaluation of SMP Clusters

Clusters of Personal Computers have been proposed as potential replacements for expensive compute servers. One limitation in the overall performance is the interconnection network. A possible solution is to use multiple processors on each node of the PC cluster. Parallel programs can then use the fast shared memory to exchange data within a node, and access the interconnection network to commun...

متن کامل

Fast Barrier Synchronization on Shared Fast Ethernet

Shared LAN is presently the most widespread networking technology, due to its extremely low cost and favourable cost/performance ratio. Clusters of Personal Computers (PCs) leveraging shared 100base-T Ethernet may currently ooer the best price/performance in parallel processing. Most numerical parallel algorithms make heavy use of collective communications and especially barrier synchronization...

متن کامل

Parallel Computation on Interval Graphs Using PC CLusters: Algorithms and Experiments

The use of PC clusters interconnected by high performance local networks is one of the major current trends in parallel/distributed computing. These clusters can yield effective parallel systems for a fraction of the price of machines using special purpose hardware. Although significant effort has been undertaken on system-level and programming environment issues for such clusters, much less at...

متن کامل

Scalability of a tera-scale linux-based clusters for parallel ab initio molecular dynamics applications

KISTI supercomputing center has initiated a TeraCluster project to build a linux-based cluster with tera-flops performance. The main goal of the project is to provide a resources composed of PC clusters that meet the level of computing power required by the grand challenge applications in Korea. In the beginning of 2002, we have built a prototype of TeraCluster with 128 computing nodes, Phase-I...

متن کامل

Turbulent Flow and Aeroacoustic Simulations using a Cluster of Workstations

The COst effective COmputing Array (COCOA), and its successor COCOA-2, is a PC cluster used for a wide variety of aerospace applications. COCOA is a 50 processor Linux cluster of off-the-shelf PCs (dual PII-400s) with 12.5 GB memory connected via fast-ethernet (100 Mbit/sec) and was built in 1998. COCOA-2 is a 42 processor rack-mounted cluster (dual PIII-800s) with 21 GB memory connected via tw...

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:

دوره   شماره 

صفحات  -

تاریخ انتشار 1999