Download presentation
Presentation is loading. Please wait.
1
Cloud Resilience System
By: Arij Alfaidi
2
Main Goal To make the “context switching time” between two VMs shortest. If one server is not working the other server should cover it up. 2/19/2019 aalfaidi/CRS
3
Main Idea Adopting mirror processes (cloud service) on both Virtual Machines, and context switcher process (cloud server), splitting large file R/W operations into small sized cloud block R/W operations. 2/19/2019 aalfaidi/CRS
4
Cloud Server ( Data Broker )
Cloud Service Cloud Service Cloud Server ( Data Broker ) Cloud Client Client PC 2/19/2019 aalfaidi/CRS
5
Main commands supported by the cloud client
NO Cloud Client Description 1 Help List supported shell commands and their usage 2 List List filenames that the cloud system kept If a special filename is given as parameter, its size will be returned 3 Get Download a file from the cloud system to local storage of the client machine 4 Put Upload a file to the cloud system from the local storage of the client machine 5 Quit Quit remote file manager shell 2/19/2019 aalfaidi/CRS
6
Data flow diagram of the Cloud system
LIST, INFO, READ, WRITE request (Type, Size, Offset, Filename) Client Shell > Request Relay the requests, resort the results and return it to the client If INFO/READ operation returned, skip other cloud machines. Cloud machine IP list Server Accept the request and treat it with file operations Return result or error code if fails Request Request Service1 Service2 2/19/2019 aalfaidi/CRS
7
Request packet data structure shared in the cloud system
Type Size Offset Path name LIST : ‘L’ INFO: ‘I’ GET : ’R’ PUT : ’W’ 4096 8 <= 4096 As file operation “empty” File name 1B 4B 8B Max? 240B 2/19/2019 aalfaidi/CRS
8
Implementation constraints
Cloud client, server, and services are using TCP socket for safe communication. Each socket client has time constraints about 2~6 seconds. 2/19/2019 aalfaidi/CRS
9
Testing 2/19/2019 aalfaidi/CRS
10
References Abhinav Mishra and Anil Kumar, “Context Switching In Clouds: Analysis And Enhancement”, International Journal of Engineering Research & Technology (IJERT) Vol. 2 Issue 7, July David Linthicum, “Build a cloud redundancy strategy using multiple providers” Luqun Li, “An Optimistic Differentiated Service Job Scheduling System for Cloud Computing Service Users and Providers”, 2009 IEEE, DOI /MUE Reese, G., “Cloud Application Architectures: Building Applications and Infrastructure in the cloud (Theory in Practice)”, O„Reilly Media, 1st Ed., 2009. Xing Pu, Ling Liu, YiduoMei, SankaranSivathanu, YounggyunKoh, CaltonPu, “Understanding Performance Interference of I/O Workload in Virtualized Cloud Environments”, 2010 IEEE DOI /CLOUD 2/19/2019 aalfaidi/CRS
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.