A New Algorithm for Solving Large-Scale Generalized Eigenvalue Problem Based on Projection Methods

Table 2

The minimum relative residual of Algorithms 1–4 in Example 2.

N.A, N.B	m = 16, N = 32		M = 18, N = 32 m = 16	M = 32, m = 4, l = 4		M = 4, L = 4, N = 32
				Threshold =
	SS	NE	CIRR	BSS	NE	BSSRR	NE
n = 200	0.9307	16	—	9.919E − 05	16	6.1079E − 16	16
n = 400	0.89555	16	—	1.3066E − 04	16	8.3165E − 16	16
n = 600	0.8926	16	—	6.6731E − 004	16	9.03555E − 16	16
n = 800	0.8802	16	—	1.600E − 004	16	8.6891E − 16	16
n = 1000	0.8749	16	—	3.2000–004	16	1.2295E − 15	16
n = 1500	0.8432	16	—	1.6400E − 004	16	1.9151E − 15	16