Performance analysis on multi-attribute combinatorial double auction model for resource allocation in cloud computing
Cloud computing is a distinctive form of the recent well-developed distributed computing which supplies multiple services to the customers on their demand. Recently, the main concern of cloud computing is a typical resource management, especially in terms of resource allocation. Various number...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2017
|
| Subjects: | |
| Online Access: | http://psasir.upm.edu.my/id/eprint/67856/1/FSKTM%202017%2020%20IR.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Cloud computing is a distinctive form of the recent well-developed distributed
computing which supplies multiple services to the customers on their demand.
Recently, the main concern of cloud computing is a typical resource management,
especially in terms of resource allocation. Various number of methods were and still
being proposed by researchers in order to provide sufficient solutions that overcome
the issues of current resource allocation methods. In this work, a performance analysis
is conducted on a dynamic market based algorithm for resource allocation in virtual
machines of the cloud. For multi-attribute combinatorial double auction model where
the simulation experiments was performed to simulate the actual business auctions’
procedures in order to consider the profits for both the cloud customers and providers,
manage the QoS metrics that being provided to cloud customers, and apply penalties
on false QoS providers as well as compensating the customers. The results showed
that multi-attribute combinatorial double auction model has enhanced the previous combinatorial double auction resource allocation model by including QoS in
provider’s bids, prevented SLA violation by penalty imposition and guaranteed
customers’ satisfaction with delivered service. And for further analysis two more
parameters were measured which are execution time and VMs’ utilization was improved. |
|---|