A Credits Based Scheduling Algorithm with K-means Clustering

Cloud Computing offers utility-oriented IT services to users allowing them to pay as they need. The key enabling factor for cloud computing is the virtualization technology [4] which virtually creates many instances of the operating systems and manages various applications on the available computing...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2018 First International Conference on Secure Cyber Computing and Communication (ICSCCC) s. 82 - 86
Hlavní autoři: Sharma, Vrajesh, Bala, Manju
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.12.2018
Témata:
Cloud computing
Clustering algorithms
Clustering of jobs
Credit Based Scheduling
K-means
Load Balancing
QoS
Quality of service
Resource management
Scheduling Algorithm based on Credits and K-means
Scheduling algorithms
Task analysis
Virtual machining
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Cloud Computing offers utility-oriented IT services to users allowing them to pay as they need. The key enabling factor for cloud computing is the virtualization technology [4] which virtually creates many instances of the operating systems and manages various applications on the available computing resources. A great challenge in cloud computing is optimal resource management to improve the utilization. To overcome the shortcomings of the existing system and to improve the efficiency of the cloud platform a novel arrangement has been insinuated. We have proposed an optimized scheduling algorithm which uses K-Means Clustering Algorithm for grouping of tasks and Virtual Machines (VM). The proposed system categorizes the virtual environment based on the available applications in each machine. For the implementation of the same, CloudSim 3.0.3 simulator has been used with the framework of Java. It has been observed that the proposed Scheduling Algorithm based on Credits and K-means clustering, has outperformed other job scheduling algorithms and it is empirically supported by experimental results as well.
DOI:10.1109/ICSCCC.2018.8703201