TY - GEN
T1 - Improving performance of multiple sequence alignment analysis in multi-client environments
AU - Catalyurek, U.
AU - Stahlberg, E.
AU - Ferreira, R.
AU - Kurc, T.
AU - Saltz, J.
N1 - Publisher Copyright:
© 2002 IEEE.
PY - 2002
Y1 - 2002
N2 - This paper is concerned with the efficient execution of multiple sequence alignment methods in a multiple client environment. Multiple sequence alignment (MSA) is a computationally expensive method, which is commonly used in computational and molecular biology. Large databases of protein and gene sequences are available to the scientific community. Oftentimes, these databases are accessed by multiple users to execute MSA queries. The data server has to handle multiple concurrent queries in such situations. We look at the effect of data caching on the performance of the data server. We describe an approach for caching intermediate results for reuse in subsequent or concurrent queries. We focus on progressive alignment-based strategies, in particular the CLUSTAL W algorithm. Our results for 350 sets of sequences show an average speedup of up to 2.5 is obtained by caching intermediate results. Our results also show that the cache-enabled CLUSTAL W program scales well on a SMP machine.
AB - This paper is concerned with the efficient execution of multiple sequence alignment methods in a multiple client environment. Multiple sequence alignment (MSA) is a computationally expensive method, which is commonly used in computational and molecular biology. Large databases of protein and gene sequences are available to the scientific community. Oftentimes, these databases are accessed by multiple users to execute MSA queries. The data server has to handle multiple concurrent queries in such situations. We look at the effect of data caching on the performance of the data server. We describe an approach for caching intermediate results for reuse in subsequent or concurrent queries. We focus on progressive alignment-based strategies, in particular the CLUSTAL W algorithm. Our results for 350 sets of sequences show an average speedup of up to 2.5 is obtained by caching intermediate results. Our results also show that the cache-enabled CLUSTAL W program scales well on a SMP machine.
UR - https://www.scopus.com/pages/publications/84966501422
U2 - 10.1109/IPDPS.2002.1016584
DO - 10.1109/IPDPS.2002.1016584
M3 - Conference contribution
AN - SCOPUS:84966501422
T3 - Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2002
SP - 183
BT - Proceedings - International Parallel and Distributed Processing Symposium, IPDPS 2002
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 16th International Parallel and Distributed Processing Symposium, IPDPS 2002
Y2 - 15 April 2002 through 19 April 2002
ER -