An Efficient MapReduce-Based Parallel Clustering Algorithm for Distributed Traffic Subarea Division

Table 2

Execution time comparison on seven data sets.

Data sets	Size (MB)	Execution time (s)
		-Means	ParCLARA	Par2PK-Means	Par3PKM
Taxi Trajectory	80	75	662	439	312
	160	103	756	486	371
	320	195	893	579	406
	640	1125	1173	838	614
	1280	2373	1679	1348	1026
	2560	—	2721	2312	1879
Iris	80	32	612	301	213
	160	54	693	380	279
	320	116	812	440	306
	640	685	1045	630	403
	1280	1800	1248	1005	768
	2560	—	2463	2013	1420
Haberman's Survival	80	56	576	311	296
	160	60	675	400	324
	320	130	823	470	378
	640	720	987	670	426
	1280	2010	1321	1200	873
	2560	—	2449	2200	1719
Ecoli	80	38	628	330	283
	160	66	712	400	324
	320	130	835	460	375
	640	756	1104	700	482
	1280	1912	1636	1234	763
	2560	—	2479	2312	1416
Hayes-Roth	80	41	568	310	278
	160	57	643	395	347
	320	125	726	460	387
	640	715	973	660	438
	1280	1980	1479	1211	736
	2560	—	2423	2120	1567
Lenses	80	32	624	309	297
	160	59	701	389	327
	320	130	924	452	376
	640	700	1072	545	432
	1280	1895	1378	1085	814
	2560	—	2379	2089	1473
Wine	80	50	635	350	317
	160	78	705	420	356
	320	130	835	470	402
	640	730	1006	610	426
	1280	2100	1346	1245	843
	2560	—	2463	2240	1645