Figure 6
From: Optimising parallel R correlation matrix calculations on gene expression data using MapReduce
Performance on the micro-benchmark. This is the performance evaluation using vanilla R, default Snowfall package with socket connection, optimised Snowfall with Rmpi package and RHIPE package with the basic MapReduce algorithm and our optimised one. The bottom three bars in each method shows the data preparation time. The vanilla R data preparation indicates loading data from a local file into memory; while in all parallel R methods, data copy almost occupy the whole data preparation time. The upper three bars respectively indicate the Euclidean (E), Pearson (P) and Spearman (S) calculation time.