Rapid advances in computer communication technology will soon enable a global information network that will have an unprecedented impact on our daily lives. Two key issues will need to be addressed for this to be successful: integration of services, and guarantees of service quality. High speed networks will carry audio, video, and data traffic from many diverse applications; unlike traditional networks, they will need to provide guarantees of Quality of Service (QoS) to applications, as required. However, application-level QoS requirements are best specified in terms meaningful to the application, while network level QoS requirements are best specified in terms meaningful to the network. The problem is to map between the two. This problem is further exacerbated by the existence of multiple intermediate protocol layers between the application layer and the network layer, as well as the specification of service quality using different parameters at the various layers. Thus a key challenge in the deployment of high speed networking systems is the understanding and development of accurate QoS transformation algorithms that request the minimal network resources to satisfy a specific application requirement. Without such an integrated, holistic treatment of the QoS issues, application users will not see the projected performance increases and may end up requesting more network resources than necessary. This research project will address several fundamental questions on application performance in high speed networks that provide QoS guarantees, and study the relationship between network level QoS and application level QoS in general. Results obtained by simulation and theoretical analysis will be validated empirically. The education component of the career development plan will include mentoring activities for women and minority graduate and undergraduate students with the objective of encouraging them to pursue research and higher education in the networking and communications area. Further, a new course on QoS issues for applications will be developed with the first offering as a research seminar. A final component of the education plan will be to develop innovative laboratory exercises on wide area networking technologies to enhance classroom understanding of concepts. WWW: www.pitt.edu/~sujata

Agency
National Science Foundation (NSF)
Institute
Division of Computer and Network Systems (CNS)
Type
Standard Grant (Standard)
Application #
9702389
Program Officer
Admela Jukan
Project Start
Project End
Budget Start
1997-07-01
Budget End
2002-06-30
Support Year
Fiscal Year
1997
Total Cost
$205,000
Indirect Cost
Name
University of Pittsburgh
Department
Type
DUNS #
City
Pittsburgh
State
PA
Country
United States
Zip Code
15213