Multiprocessing MCNP on an IBN RS/6000 cluster
- Los Alamos National Lab., NM (United States)
- IBM Corp., Houston, TX (United States)
The advent of high-performance computer systems has brought to maturity programming concepts like vectorization, multiprocessing, and multitasking. While there are many schools of thought as to the most significant factor in obtaining order-of-magnitude increases in performance, such speedup can only be achieved by integrating the computer system and application code. Vectorization leads to faster manipulation of arrays by overlapping instruction CPU cycles. Discrete ordinates codes, which require the solving of large matrices, have proved to be major benefactors of vectorization. Monte Carlo transport, on the other hand, typically contains numerous logic statements and requires extensive redevelopment to benefit from vectorization. Multiprocessing and multitasking provide additional CPU cycles via multiple processors. Such systems are generally designed with either common memory access (multitasking) or distributed memory access. In both cases, theoretical speedup, as a function of the number of processors P and the fraction f of task time that multiprocesses, can be formulated using Amdahl's law: S(f, P) =1/(1-f+f/P). However, for most applications, this theoretical limit cannot be achieved because of additional terms (e.g., multitasking overhead, memory overlap, etc.) that are not included in Amdahl's law. Monte Carlo transport is a natural candidate for multiprocessing because the particle tracks are generally independent, and the precision of the result increases as the square Foot of the number of particles tracked.
- OSTI ID:
- 5874672
- Report Number(s):
- CONF-930601-; CODEN: TANSAO
- Journal Information:
- Transactions of the American Nuclear Society; (United States), Vol. 68; Conference: American Nuclear Society (ANS) annual meeting, San Diego, CA (United States), 20-24 Jun 1993; ISSN 0003-018X
- Country of Publication:
- United States
- Language:
- English
Similar Records
Multiprocessing MCNP on an IBM RS/6000 cluster
Distributed processor Monte Carlo: MCNP results on a 16-node IBM cluster