Difference between revisions of "Fetahi Z. Wuhib"

From SimpleWiki
Jump to navigationJump to search
Line 61: Line 61:
  
 
== External links ==
 
== External links ==
* [http://www.ee.kth.se/~fzwuhib/] Homepage of Fetahi Z. Wuhib
+
* [http://www.ee.kth.se/~fzwuhib/ Homepage ] of Fetahi Z. Wuhib
* Publications of Fetahi Z. Wuhib, as [http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/_XXXX_ indexed by DBLP]
+
* Publications of Fetahi Z. Wuhib, as [http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/w/Wuhib:Fetahi.html indexed by DBLP]
  
 
[[Category:PhD students]]
 
[[Category:PhD students]]
 
[[Category:People]]
 
[[Category:People]]

Revision as of 13:18, 20 May 2010

Summary
Student: Fetahi Z. Wuhib
Title: A self-organizing, decentralized management

layer for real-time monitoring

e-mail: fetahi@kth.se
Affiliation: KTH
Supervisor: Rolf Stadler
Committee:
Start: 2005
End: 2009
Funding: The Swedish Foundation for Strategic Research and

The Graduate School of Telecommunications

Biography

Fetahi Zebenigus Wuhib is a Ph.D. candidate at the Royal Institute of Technology (KTH), Stockholm, Sweden. He received his B.Sc. degree in electrical engineering from Addis Ababa University, Ethiopia, his M.Sc. degree in internetworking and his Licentiate of Technology degree in telecommunications from KTH. His current research focuses on distributed protocols for network monitoring and configuration.

PhD project description

The goal of this research is to engineer scalable and fault-tolerant management mechanisms to manage large-scale dynamic systems.

In this context, we design and evaluate a management layer that monitors aggregates of management data from devices across the network in near real-time. Aggregation functions that are often used by management applications include SUM, MAX and AVERAGE of device-level counters. Our approach is to engineer a self-organizing management overlay, which can be seen as a thin layer of functionality that interconnects all devices/functional entities in a network environment. Such a system is completely decentralized with all nodes having identical functionality. The advantage of such a design is scalability and robustness. The monitoring functions will be based on distributed algorithms that execute on the management overlay. We evaluate different classes of distributed algorithms regarding their suitability for real-time monitoring and further develop them if needed.

We also design and evaluate a resource management system for a cloud environment. A cloud is composed of a set of machines and a set of applications that run on the machines. The problem is to design and evaluate a management system that fairly allocates resources to the applications in a scalable and robust manner. Our approach is to use distributed algorithms to achieve our goals.

References

  1. F. Wuhib M. Dam R. Stadler and A. Clemm. Decentralized computation of threshold crossing alerts. In Proc. 16th IEEE/IFIP Workshop on Distributed Systems: Operations and Management (DSOM), pages 220–232, Barcelona Spain, October 24-26, 2005.
  2. F. Wuhib, R. Stadler and A. Clemm, "Implementation and Evaluation of a Protocol for Detecting Network-Wide Threshold Crossing Alerts", 4th IFIP/IEEE Workshop on End-to-End Monitoring Techniques and Services, Vancouver, Canada, April 3 2006.
  3. F. Wuhib, R. Stadler and A. Clemm, "Decentralized Service Level Monitoring using Network Threshold Crossing Alerts", IEEE Communications Magazine, Vol. 44, Issue 10, October 2006.
  4. F. Wuhib, M. Dam, R. Stadler and Alexander Clemm, "Robust Monitoring of Network-wide Aggregates through Gossiping", 10th IFIP/IEEE International Symposium on Integrated Network Management (IM 2007), Munich, Germany, May 2007.
  5. F. Wuhib, "Real-time Monitoring of Global Variables in Large-scale Dynamic Systems", KTH Licentiate Thesis, TRITA-EE 2007:065, December 2007.
  6. F. Wuhib, M. Dam and R. Stadler, "Decentralized Detection of Global Threshold Crossings Using Aggregation Trees, Computer Networksl, Vol. 52, pp 1745-1761, February 2008, doi:10.1016/j.comnet.2008.02.015.
  7. F. Wuhib, M. Dam, R. Stadler and A. Clemm, "Robust Monitoring of Network-wide Aggregates through Gossiping", IEEE Transactions on Network and Service Management, Vol. 6, no. 2, June 2009.
  8. F. Wuhib, M. Dam and R. Stadler, "Gossiping for Threshold Detection", 11th IFIP/IEEE International Symposium on Integrated Network Management (IM 2009), Long Island, NY, June 1-5, 2009.
  9. F. Wuhib, M. Dam and R. Stadler, "A Gossiping Protocol for Detecting Global Threshold Crossings", IEEE transactions on Network and Service Management, Vol 7. no. 1, March 2010.
  10. S. Krishnamurthy, J. Ardelius, E. Aurell, M. Dam, R. Stadler and F. Wuhib, "The Accuracy of Tree-based Counting in Dynamic Networks", February 2010, Submitted.

Additional information

External links