Grant: $200,000 - National Science Foundation - Aug. 1, 2009
No votes have been cast for this award yet
Award Description: NeTS: Medium: Collaborative Research: Towards Versatile and Programmable Measurement Architecture for Future Networks. This project envisages and advocates a versatile measurement architecture as an integral part of any future network substrate, with not only built-in hardware and software primitives incorporated into network elements, but also network-wide architectural support for network measurement. Such a measurement architecture must be adaptive, versatile, programmable, modular, and scalable so as to meet the complex, diverse, and fluid needs of future network services and applications--many of which are yet to be imagined. We are exploring various architectural issues in realizing such vision, and seek to answer important research questions, e.g., what constitutes primitive queries, what semantics are needed to compose meaningful measurement tasks based on these primitives, and what level of flow abstractions are required. These are fundamental yet challenging problems that must be addressed to develop a unified architecture that can leverage recent advancement in sampling theory, data streaming, and reconfigurable hardware platforms. To answer these questions, our research tasks are organized along on the following closely related themes: 1) Design of a versatile and programmable architecture with flexible semantics, built-in capabilities and network-wide support to enable network operators and other users to specify measurement goals and compose specific measurement tasks based on the underlying hardware and software primitives. 2) Development of powerful flow abstraction module with a powerful notion of flowset as the base unit of network measurement and a new flowset composition language (FCL) that allows for arbitrary definitions of traffic subpopulation and/or aggregates of interest for measurement purposes. 3a) Development of intelligent factorization module that decomposes complex measurement tasks and queries into a set of efficient canonical 3b) measurement primitives such as sampling and data streaming operations. 4) Development of scalable and composable hardware primitives that efficiently implement the set of measurement primitives and enables the collection of high-fidelity data for different ``flowsets.'' 5) Measurement application case studies will also be carried out both to guide and fine-tune the design of the proposed measurement architecture as well as to evaluate and showcase its capabilities.
Project Description: As defined in the award description field
Jobs Summary: No jobs reported (Total jobs reported: 0)
Project Status: Less Than 50% Completed
This award's data was last updated on Aug. 1, 2009. Help expand these official descriptions using the wiki below.