Skip to Main Content
Probably the single most important concern of the biotechnology industry is to improve existing biotechnology applications and tools due to the exponential increase in the size of the datasets. Improving application specific performance, pertaining to sudden and dynamic changes in the execution environment has been a widely researched problem. This research was undertaken to find application specific performance problems in the field of applied Bioinformatics. We specifically narrow down on an important everyday bioinformatics application namely Blast, used commonly by biologists for everyday research. A detailed literature study was undertaken to observe the gaps specific to the applications performance subjected to different quality of service constraints and system specific parameters under the confluence of a dynamic mini grid test bed. A number of gaps were identified; throughput in terms of latency (message passing); scalability (load balancing) and application performance (reliability) needs to be addressed. An experimental mini grid test bed was implemented to simulate various conditions and to test the proposed hypotheses. The implemented architecture utilizes game theoretic optimization and agent based team formation (Coalition) algorithms to improve upon scalability with respect to team formation. Due to the dynamic nature of distributed systems (as discussed in previous works) all interactions are made local within a team transparently. This paper is a proof of concept of an experimental mini-Grid test-bed aimed at improving performances of bioinformatics Blast application in terms of scalability and stability. Experimental results and detailed literature on current approaches are explored to conduct a set of generic experiments to validate our claims.
Date of Conference: June 29 2008-July 5 2008