IBM Israel
Skip to main content
 
Search IBM Research
   Home  |  Products & services  |  Support & downloads  |  My account
Select a Country Select a country
IBM Research Home IBM Research Home
IBM Haifa Labs Homepage IBM Haifa Labs Home

Cluster Virtual Machine for Java

Distributed Computing Systems
Project Homepage
 ·Overview
 ·Motivation
 ·Application Model
 ·Approach
 ·Benchmark Results
 ·Optimizations
 ·Status
 ·Related Works
 ·Papers
 ·Contact Information
Feedback


Benchmark Results
The following graph shows the efficiency/speedup of Cluster VM for Java on the applications described below:



The benchmarks used are:
  • N-Body
    Simulates the motion of n particles due to gravitational forces over ts number of simulation time steps, using m threads. n = 640, m = 4 and ts = 10 were used.

    The program follows the algorithm of Barnes & Hut in "A Hierarchical O(n log n) Force Calculation Algorithm" J. Barnes and P. Hut, Nature, v. 324(4), pages 446--449, December 1986.

    The N-Body benchmark

  • pBOB
    Implements a TPC-C inspired business logic kernel. Creates m threads that generate transactions concurrently. m = 4 was used. In accordance with the TPC's fair use policy we note that pBOB deviates from the TPC-C specification and is not comparable to any official TPC results.

    A modified version of pBOB was adopted by SPEC as SPECjbb2000. SPECjbb2000 is not identical to pBOB.

  • TSP
    Find in parallel the shortest route among n cities using m threads. n = 14 and m = 4 was used.

    The algorithm follows the one described in Richard M. Karp and Yanjun Zhang, "Randomized Parallel Algorithms for Backtrack Search and Branch-and-bound Computation", Journal of the ACM, 40(3):765-789, July 1993.

    The TSP benchmark


 

  About IBM  |  Privacy  |  Legal  |  Contact