A Comprehensive Availability Modeling and Analysis of a Virtualized Servers System Using Stochastic Reward Nets

Table 3

Input parameter values used in the analysis.

Parameters	Description	Assigned transitions	Mean time/values
	Host repair rate	TH1r, TH2r	3 days
	Host failure rate	TH1f, TH2f	1 year
	VMM restart rate from downstate	TVMM1restart, TVMM2restart	1 min
	VMM nonaging failure rate	TVMM1f, TVMM2f	2654 hours
	VMM repair rate	TVMM1repair, TVMM2repair	100 mins
	VMM aging rate	TVMM1fp, TVMM2fp	1 month
	VMM aging failure rate	TVMM1af, TVMM2af	1 week
	VMM recovery rate after aging failure	TVMM1arecovery, TVMM2arecovery	65 mins
	VMM clock interval	TVMM1clockinterval, TVMM1clockinterval	1 week
	VMM rejuvenation rate	TVMM1rej, TVMM1rej	2 mins
	SAN failure rate	TSANf	1 year
	SAN repair rate	TSANrepair	3 days
	VM nonaging failure rate	TVM1f, TVM2f	2893 hours
	VM repair rate	TVM1repair, TVM2repair	30 mins
	VM restart rate	TVM1restart, TVM2restart	30 s
	VM aging rate	TVM1fp, TVM2fp	1 week
	VM aging failure rate	TVM1af, TVM2af	3 days
	VM recovery rate after aging failure	TVM1arecoveryt, TVM2arecovery	35 mins
	VM clock interval	TVM1clockinterval, TVM2clockinterval	1 day
	VM rejuvenation rate	TVM1rej, TVM2rej	1 min
	VM booting rate after VMM rejuvenation	TVM1boot, TVM2boot	30 s
c VMM	Number of stages in c VMM-stage Erlang distribution	x	2
c VM	Number of stages in c VM-stage Erlang distribution	x	2
n VM	Number of VMs running on a VMM	x	2