A Web-based and Grid-enabled dChip version for the analysis of large sets of gene expression data

BMC Bioinformatics

Table 6 Mean execution times and speedup ratio analysing 100 microarrays on cluster environment

	1 job	5 job	10 job	15 job	20 job
Module 1	285 sec	63.80 sec	35.20 sec	26.40 sec	25.65 sec
Module 2	344 sec	81.60 sec	58.00 sec	44.47 sec	39.55 sec
Module 1+2	629 sec	145.40 sec	93.20 sec	70.87 sec	65.20 sec
speedup ratio module 1		4.47	8.09	10.79	11.11
speedup ratio module 2		4.22	5.93	7.74	8.69
speedup ratio module 1+2		4.33	6.75	8.88	9.65

The table shows the mean execution times coming from three different execution of dChip MPI version run on LITBIO cluster using respectively 1, 5, 10, 15, 20 parallel jobs to compute the expression values of 100 microarrays. To estimate the parallelization efficiency, the speedup ratio (S(N) = T(1)/T(N), where T(1) is the execution time on a single processor and T(N) the execution time on N processors) was calculated.

ISSN: 1471-2105