VM placement control considering network bandwidth for high availability distributed clusters

2017 ◽  
Author(s):  
Takahiro Yada ◽  
Kiyoshi Ueda
Keyword(s):  
High Availability ◽  
Vm Placement ◽  
Network Bandwidth ◽  
Distributed Clusters

Optimum Utilization of Resources Through Restricted Virtual Machine Migration and Efficient VM Placement in Cloud Data Center

2018 ◽  
Vol 9 (4) ◽  
pp. 1-19
Author(s):  
Subhadra Bose Shaw ◽  
Anil Kumar Singh ◽  
Shailesh Tripathi
Keyword(s):  
Energy Consumption ◽  
Virtual Machines ◽  
Energy Performance ◽  
Migration Decision ◽  
Performance Gain ◽  
Virtual Machine Migration ◽  
Vm Placement ◽  
Cloud Data Center ◽  
Network Bandwidth ◽  
Optimum Utilization

In infrastructure-as-a-service (IAAS) cloud platforms, it is a real challenge to provide high performance gain by the optimum utilization of resources while maintaining minimum consumption of energy. The existing research works show that reduction in energy consumption causes violation of service level agreement (SLA). In this article, the concept of probability has been used to take the migration decision of virtual machines (VM) from over-utilized as well as under-utilized nodes. A novel method has also been proposed for selecting the destination server where a migrated VM will be placed. This method is based on the current utilization of CPU, memory and network bandwidth. The proposed scheme maintains a balance between energy consumption and performance gain. Results obtained through trace driven simulation demonstrate that the probability-based migration scheme achieves energy-performance trade-off whereas the VM placement method shows a very high gain in performance.

scholarly journals Decision Making Framework for Decentralized Virtual Machine Placement in Cloud Computing

2018 ◽  
Vol 7 (2.7) ◽  
pp. 705
Author(s):  
Suresh B.Rathod ◽  
V Krishna Reddy
Keyword(s):  
Decision Making ◽  
Virtual Machine ◽  
Dynamic Environment ◽  
Resource Usage ◽  
Vm Placement ◽  
Resource Requirement ◽  
Network Bandwidth ◽  
Decentralized Decision Making ◽  
Decision Making Framework ◽  
Host Identification

In distributed cloud environment hosts are configured with Local Resource Monitors (LRM). This LRM monitors the underlying hosts’ resource usage, runs independently and balances the underling host’s load by migrating Virtual Machine (VM) instance. For the dynamic environment, each hosts has varying resource requirement, hosts load cannot remain constant. LRM at each host takes decision for VM migration considering static threshold on its own and other hosts current CPU utilization. This result in chances of getting selected same host for VM placement by multiple hosts to reduce resource usage of underlying hosts. The decision making at each server causes the problem of same host identification by multiple hosts during VM placement and consumes extra CPU power and network bandwidth consumption towards each server. This paper addresses the above said issue by proposing decentralized decision making framework for cloud considering hybrid Peer to Peer (P2P) network topology. Proposed solution results avoiding above said issues and balances the load across servers in DC.  

HALR: A TCP Enhancement Scheme Using Local Caching in High-Availability Cluster

2009 ◽  
Vol E92-B (1) ◽  
pp. 26-33
Author(s):  
Yi-Hsuan FENG ◽  
Nen-Fu HUANG ◽  
Yen-Min WU
Keyword(s):  
High Availability

A Study for VM Placement Schemes in Cloud

2017 ◽  
Vol 7 (8) ◽  
pp. 208
Author(s):  
Ramandeep Kaur ◽  
Navpreet Kaur
Keyword(s):  
Cloud Computing ◽  
Reaction Time ◽  
Task Scheduling ◽  
Specific Time ◽  
Due Date ◽  
Vm Placement ◽  
Time Task ◽  
Wide Range ◽  
The Web

The cloud computing can be essentially expressed as aconveyance of computing condition where distinctive assets are conveyed as a support of the client or different occupants over the web. The task scheduling basically concentrates on improving the productive use of assets and henceforth decrease in task fruition time. Task scheduling is utilized to allot certain tasks to specific assets at a specific time occurrence. A wide range of systems has been exhibited to take care of the issues of scheduling of various tasks. Task scheduling enhances the productive use of asset and yields less reaction time with the goal that the execution of submitted tasks happens inside a conceivable least time. This paper talks about the investigation of need, length and due date based task scheduling calculations utilized as a part of cloud computing.

Clustering Server Pada Cloud Computing Berbasis Proxmox VE Menggunakan Metode High Availability

2018 ◽  
Vol 3 (1) ◽  
Author(s):  
Linda Apriliana ◽  
Ucuk Darusala Darusalam ◽  
Novi Dian Nathasia
Keyword(s):  
Cloud Computing ◽  
File System ◽  
High Availability ◽  
Client Server ◽  
Network File System

Layanan dan data teknologi Cloud Computing tersimpan pada server, hal ini menjadikan faktor pentingnya server sebagai pendukung ketersediaan layanan. Semakin banyak pengguna yang mengakses layanan tersebut akan mengakibatkan beban kinerja mesin server menjadi lebih berat dan kurang optimal, karena layanan harus bekerja menyediakan data terus-menerus yang dapat diakses kapanpun oleh penggunanya melalui jaringan terkoneksi. Perangkat keras server memiliki masa performa kinerja. Hal serupa dengan perangkat lunak yang dapat mengalami crash. Dengan fungsi server yang memberikan layanan kepada client, server dituntut untuk memiliki tingkat availability yang tinggi. Hal tersebut memungkinkan mesin server mengalami down. Server juga harus dimatikan untuk keperluan pemeliharaan. Penelitian bertujuan ini membangun Clustering Server yang dapat bekerja bersama yang seolah merupakan sistem tunggal diatas lingkungan virtual. Hal ini merupakan solusi untuk mengatasi permasalahan tersebut. Pada penelitian ini penulis menggunakan server virtualisasi proxmox, FreeNAS sebagai server NAS dan DRBD untuk pendukung ketersediaan layanan tinggi dalam lingkup HA, sinkronisasi data dalam High Availability (HA) yang dapat melakukan mirroring sistem kemesin lain. Dengan diterapkannya metode HA dan sinkronasi DRBD serta penggunaan NFS (Network File System) pada sistem cluster didapatkan hasil rata-rata waktu migrasi sebesar 9.7(s) pada node1 menuju node2, 3.7(s) node2 menuju node3, dan 3(s) pada node3 menuju node1. Didaptkan juga waktu downtime yang lebih sedikit yaitu sebesar 0.58 ms pada node1, 0.02 ms pada node2, dan 0.02 ms pada node3.

Sign in / Sign up

Export Citation Format

Share Document