Information Technology Reference
In-Depth Information
Assumption 2
1. The probability the task fails at state
t
i
is l (or failure rate), and
follows a Poisson distribution. Mean time to failure
(MTTF) is
mathematically dei ned by 1/l.
2. The task fails at state
t
i
, and the time when a failure occurs follows
a Poisson distribution.
3. The failure execution time of the task is
T
F
.
4. The average down time of the system is
T
D
.
Assumption 3
1. The number of replicas of a task is
N
R
for replication policy or
replication with checkpointing policy.
2. The overhead of checkpoint (
T
C
) is the same for different policies.
3. The recovery time of checkpointing (
T
R
) is the same for different
policies.
4. The checkpoint interval is
T
I
.
With these dei nitions and assumptions, a brief explanation of QoS crite-
ria for different failure-recovery policies is presented.
Retrying
The computation of the execution time with retrying was i rst specii ed by
Duda [71].
l
T
l
T
ee
(
-
1)
D
F
Q
=
(4.17)
et ret
()
l
The execution cost for retrying is computed as
Q
co
=
Q
price
¥
Q
et(ret)
(4.18)
Checkpointing
The execution time with checkpointing is computed from Equation 4.19
and adopted from [72].
ʈ
l-+
(
TTT
)
l+
(
TT
)
T
e
(
e
-
1)
DCR
I
C
(4.19)
Q
=
Á˜
F
et ch
()
˯
T
l
I
The execution cost for checkpointing is as
Q
co
=
Q
price
¥
Q
et(ch)
(4.20)
Search WWH ::
Custom Search