Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Do you really want to delete this prezi?
Neither you, nor the coeditors you shared it with will be able to recover it again.
Make your likes visible on Facebook?
You can change this under Settings & Account at any time.
Transcript of Blue Eyes
Dept. of Comput. Sci., Univ. of Maryland, College Park, MD, USA
Issue Date: 23-29 May 2009
ISSN: 1530-2075 What is Cloud Computing? Providing provisional resources on demand via a Computer Network
Hence cloud computing can provide the following
Software as a Service
Storage Companies like IBM, Microsoft are already implementing clouds Traditional Cloud Architecture Clouds provide large amounts of server space
These servers are expected to scale up depending on the particular demand
Clouds make use of the concept of virtualization
Looking ahead a more intuitive way for scaling operations is needed What is Blue Eyes? Traditional system management solutions scale to only hundreds or thousands of systems at most Blue Eyes, a new system management solution, has the following features Ability to handle hundreds of thousands of system
Uses a "scale-out" technique rather than the scale-up technique
Management servers are structured into heirarichal trees
Extends existing single server implementation without changing legacy software Blue Eyes Design Overall Architecture Traditional System Management Architecture Directly manages all end-points
Single point of failure
Unnecesarry overhead Blue Eyes Multi-Server System Management Architecture Distribution of management workload
"Detect & Switch" fail over mechanism Management Server Tree Why a Heirarchical Tree Topology? Information can be aggregated upwards & commands sent downwards Follows business organization heirarchy, ie. Management tasks locality MST is implicitly constructed using index numbers of the server
Implemented usind a list data sctructure
MST is exploited to balance the communicaion workload, ie. limited communication
Hence, eliminates worries of potential bottle-neck Management Reliability Ring A solid lined node depicts a primary server
A dashed line node depicts a backup server MRR is a logical network to provide failure resistance
It follows a primary-secondary relationship
Like the MST, MRR is also implemented via a list data structure
A special scatter placement scheme is used to allot the backup servers for better load balancing, comapred to round-robin technique Method of failure detection Primary server periodically sends a "keep-alive" and management information to its secondary/backup server Management Server List It's a globally shared data structure
Each entry contains:
Index number of a management server
2 IP addresses of the primary and secondary server
Flag that is set to P or S
During any change, the MSL is changed, and this is reflected in the MRR and the MST Ankeet Guha 1PI07IS016 Guided By: Mr. Vinayak S.P. Mechanisms Management Server Operations Server operations to maintain the distributed server network
"Add" and "Merge" operations are used to expand the network
Number of endpoints is balanced over servers when there is a change in the network "Balance", "fail-over" and "recovery" operations are managed implicitly Add Operation Merge Operation Management Tasks 5 Basic Management Tasks
event_action Discover Operation Conclusion The Blue Eyes architecture effectively shows the following:
Manage multiple servers
MST enables for efficient forwarding of instructions
MRR provides high reliability
Tests have also shown the percentage gains of this technology Thank You! References  D. Watts, et al., “Implementing IBM Director 5.20. SG24- 6188”, IBM Corp., http://publibb. boulder.ibm.com/Redbooks.nsf/RedbookAbstracts/sg246188.html, April 2007.
 Microsoft Corp., “Microsoft System Management Server
Planning Guide”, http://www.microsoft.com/technet/prodtechnol/sms/sms2/prod
 W. Zhang, “Linux Virtual Server for Scalable Network Services”, Ottawa Linux Symposium 2000.
 S. Ratnasamy, et. al., “A Scalable Content Addressable Network”, Proceedings of the ACM SIGCOMM, August 2001.