We consider a server farm with different classes of arriving requests. For each class, the request arrival rate, the mean service time and the variance of the service time are assumed to be known. The arrivals are Poisson and the service times are distributed arbitrarily. Servers are modeled a s single server queuing systems. The problem is to determine the assignment of requests’ classes to servers in order to minimize the overall expected waiting time, overall probability of the wait exceeding a given value, and related measures.
We optimize under two separate conditions. The first is the light traffic condition, and the second is the balanced load condition. In each case we obtain optimal solution for the two-server case and then based on the insights obtained we design algorithms for the multi-server case. We attempt both {\em fractional} and {\em integral} versions of the problem and show that if the number of streams is large as compared to the number of servers, then fractional and integral optimal solution values are close to each other.
The work has applications in the setting of web-server farms and E-businesses. For example, it can be used in assigning E-businesses to servers on web-server farms, or assigning customers belonging to different QoS class to different servers. Experimental results using data from actual traces show significant improvement over naive and greedy methods.
By: Rahul Garg, Perwez Shahabuddin, Akshat Verma
Published in: RI03004 in 2003
LIMITED DISTRIBUTION NOTICE:
This Research Report is available. This report has been submitted for publication outside of IBM and will probably be copyrighted if accepted for publication. It has been issued as a Research Report for early dissemination of its contents. In view of the transfer of copyright to the outside publisher, its distribution outside of IBM prior to publication should be limited to peer communications and specific requests. After outside publication, requests should be filled only by reprints or legally obtained copies of the article (e.g., payment of royalties). I have read and understand this notice and am a member of the scientific community outside or inside of IBM seeking a single copy only.
Questions about this service can be mailed to reports@us.ibm.com .