Three Strategies for Improving Shortest Vector Enumeration Using GPUs

Table 1

Average execution time in seconds for our implementations and Hermans’s implementation [28].

Hermans et al. [28] with	Dimension
	80	90	100	110
I1: Sections 5.2, 5.1.1, and 5.1.2	287	1450	4876	30,726
I2: Sections 5.2 and 5.1.1	365	2177	6305	40,836
I3: Sections 5.2	242	1764	9360	59,482
I4: —	297	2398	12,125	77,982
Speedup of I1 compared to I4	1.03	1.65	2.49	2.54
Speedup of I2 compared to I4	0.81	1.1	1.92	1.91
Speedup of I2 compared to I3	0.66	0.81	1.48	1.46
Speedup of I3 compared to I4	1.23	1.36	1.30	1.31