(Georgia Institute of Technology, 2006-02-25)
Bader, David A.; Sachdeva, Vipin
The maximum flow problem is a combinatorial problem of significant importance in a wide variety
of research and commercial applications. It has been extensively studied and implemented
over the past 40 years. The push-relabel method has been shown to be superior to other methods,
both in theoretical bounds and in experimental implementations. Our study discusses the implementation
of the push-relabel network flow algorithm on present-day symmetric multiprocessors
(SMP's) with large shared memories. The maximum flow problem is an irregular graph problem
and requires frequent fine-grained locking of edges and vertices. Over a decade ago, Anderson and
Setubal implemented Goldberg's push-relabel algorithm for shared memory parallel computers;
however, modern systems differ significantly from those targeted by their implementation in that
SMP's today have deep memory hierarchies and different performance costs for synchronization
and fine-grained locking. Besides our new cache-aware implementation of Goldberg's parallel
algorithm for modern shared-memory parallel computers, our main new contribution is the first
parallel implementation and analysis of the gap relabeling heuristic that runs from 2.1 to 4.3 times
faster for sparse graphs.
(Georgia Institute of Technology, 2006)
Bader, David A.; Li, Yue; Li, Tao; Sachdeva, Vipin
The exponential growth in the amount of genomic data
has spurred growing interest in large scale analysis of
genetic information. Bioinformatics applications, which
explore computational methods to allow researchers to sift
through the massive biological data and extract useful
information, are becoming increasingly important
computer workloads. This paper presents BioPerf, a
benchmark suite of representative bioinformatics
applications to facilitate the design and evaluation of high-performance
computer architectures for these emerging
workloads. Currently, the BioPerf suite contains codes
from 10 highly popular bioinformatics packages and
covers the major fields of study in computational biology
such as sequence comparison, phylogenetic reconstruction,
protein structure prediction, and sequence homology &
gene finding. We demonstrate the use of BioPerf by
providing simulation points of pre-compiled Alpha
binaries and with a performance study on IBM Power
using IBM Mambo simulations cross-compared with Apple
G5 executions.
The BioPerf suite (available from www.bioperf.org)
includes benchmark source code, input datasets of various
sizes, and information for compiling and using the
benchmarks. Our benchmark suite includes parallel codes
where available.
(Georgia Institute of Technology, 2006)
Bader, David A.; Sachdeva, Vipin
In this paper, we propose BIOPERF, a definitive
benchmark suite of representative applications from the biology
and life sciences community, where the codes are carefully
selected to span a breadth of algorithms and performance characteristics.
The BIOPERF suite is available from www.bioperf.
org and includes benchmark source code, input datasets of
various sizes, and information for compiling and using the
benchmarks. We include parallel codes where available.