Information Technology Reference
In-Depth Information
(a)
(b)
(c)
(d)
Fig. 3. Case Study 2: (a) Total energy consumption by data centers (b) Number of
VM migrations (c) Execution time of tasks with 2 overloaded Data Centers (d) Average
SLA Violation for the federated approach
Tabl e 1 . Comparative summary of Cloud server consolidation strategies
Paper Target Power-Aware Federated Multi-agent Migration SLA
[4]
Cloud
No
No
Yes
No
No
[6]
Cluster
No
No
Yes
No
No
[5]
Cluster
Yes
No
Yes
No
Yes
[12]
Cluster
Yes
No
No
No
Yes
[14]
Cloud
Yes
No
No
No
No
[3]
Cloud
Yes
No
No
Same DC Yes
This work Cloud
Yes
Yes
Yes
Among DCs Yes
two execute in cluster computing environments. Three approaches [12, 14, 3]
reduce power consumption in cloud computing considering SLAs. None of the
analyzed proposals consider federation cloud environment nor VM migration
among data centers.
7 Final Consideration and Future Work
In this paper, we proposed and evaluated a server consolidation approach for ef-
ficient power management in virtualized data centers taking into account energy
consumption, and QoS requirements. The results obtained in the CloudSim [3]
simulator, with 2 data centers and 400 simultaneous virtual machines show that
very good energy consumption savings are obtained with our approach, while
meeting the QoS requirements. The best gain (53.57%) was obtained when we
have one overloaded data center. In this case, we were able to reduce the energy
