SparkGC: Spark based genome compression for large collections of genomes

BMC Bioinformatics

Table 2 Runtime of different parts on different numbers of worker nodes

Chromosome	Stage	1 worker node		2 worker nodes		3 worker nodes		4 worker nodes
Chromosome	Stage	Time (s)	%	Time (s)	%	Time (s)	%	Time (s)	%
Chr1	Pre-processing	112	1.79	116	3.84	124	5.27	126	6.49
	First-order	5577	89.23	2618	86.72	2007	85.37	1648	84.81
	Second-order	531	8.50	254	8.41	188	8.00	136	7.00
	Post-processing	30	0.48	31	1.03	32	1.36	32	1.70
	Total	6250	100	3019	100	2351	100	1943	100
Chr13	Pre-processing	62	3.58	70	6.42	70	9.06	70	10.74
	First-order	1520	87.76	921	84.50	625	80.85	512	78.53
	Second-order	120	6.93	69	6.33	47	6.08	39	5.98
	Post-processing	30	1.73	30	2.75	31	4.01	31	4.75
	Total	1732	100	1090	100	773	100	652	100

ISSN: 1471-2105