Bill Lin - US grants

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5).

Traffic measurement is central to network operation, management, and security. Yet, support for measurement was not an integral part of the original Internet architecture. This project aims to develop a programmable measurement architecture that is versatile enough to support current and future measurement needs, adaptable to dynamic network conditions, modular/lightweight, and scalable to high link speeds.

This project proposes a new flow abstraction module and query language that can specify arbitrary traffic sub-populations for statistics collection. Efficient data structures to encode these queries will be developed. The team also strives to identify a core set of data streaming and sampling primitives that can be composed together to satisfy most of the queries. Efficient hardware implementation for these core set of primitives will constitute the basic measurement modules that can be easily reconfigured to measure traffic at different desired granularity. Measurement application case studies will be carried out to evaluate and showcase the capabilities of the proposed approach.

This project has great potential in guiding the design of a clean-slate measurement instrumentation for future Internet. It will provide both graduate and undergraduate students with training that span multiple disciplines, from fundamental statistical theory, algorithm design, to hardware implementation. The results (including the query language and underlying data structures, sampling/streaming algorithms, and hardware building blocks) will be broadly disseminated through publications, invited talks/tutorials, and open-sourcing software distribution.

Networks-on-chip (NoCs) are an on-chip interconnection fabric of choice for general-purpose chip multiprocessors and application-specific multiprocessor systems-on-chip. This project focuses on network-on-chip modeling and optimization to enable early-stage design exploration that fully elaborates the achievable envelope of NoC power, area, speed, reliability and cost. An architectural estimation thrust develops new architecture-level estimation methods that are portable to different router microarchitectures and that can accurately capture implementation effects. The thrust addresses the automatic generation of architectural estimation models, the modeling of chip design implementation flow choices and their impacts, and new trace-aware and workload-dependent estimations. An architectural optimization thrust develops new methodologies for trace-driven optimization of router configurations, packet routing and network topology, with consideration of runtime network resource contentions.

Successful completion of this project will help network-on-chip and multiprocessor system-on-chip designers reduce design effort while improving chip area, delay and power metrics. This will be enabling to the efficient design of more complex, higher-functionality integrated-circuit products within given cost and power limits. The research will also produce software tools to establish a foundation for further work by other research groups. Through research participation, both graduate and undergraduate students will be trained in this emerging aspect of system-on-chip design.

This INSPIRE award is partially funded by Research in Networking Technology and Systems Program in the Division of Computer and Network Systems in the Directorate for Computer and Information Science and Engineering, the Manufacturing Enterprise Systems Program in the Division of Civil, Mechanical and Manufacturing in the Directorate for Engineering, and the Communications and Information Foundations Program in the Division of Computing and Communications Foundations in the Directorate for Computer and Information Science and Engineering.

This project addresses common problems across two traditionally separate disciplines of manufacturing and networking by focusing on the following applications in the two fields-advanced semiconductor manufacturing and virtualized data center networking. Research in each field has heretofore focused on different problems, with semiconductor manufacturing largely focused on throughput maximization, mean cycle time minimization and system stability and cloud networking on network performance guarantees. However, the two areas would benefit from sharing a common integrated focus and a unifying stochastic processing network model for modeling and analyzing problems. This project contributes this new mathematical foundation along with a set of practical service disciplines and scheduling algorithms to enable reasoning about and to provide performance guarantees in stochastic processing networks. In particular, this project will concurrently investigate the problems of delivery guarantees in semiconductor manufacturing and network performance guarantees in virtualized data centers so that the two application domains can inform each other and by doing so develop new solutions that might not otherwise be imagined.

The central contribution of the proposed work will be a new mathematical foundation, which the principal investigators call Stochastic Processing Calculus that will allow researchers and practitioners to reason about and provide performance guarantees for diverse range of applications that can be modeled as stochastic processing networks. Scientific discoveries often happen at the intersection of two disciplines. This project involves very competent and accomplished researchers (in the areas of networking, network theory, and industrial systems engineering and operations research) and crosses diverse disciplines with the intension of looking at a set of intersections as it explores and advances Stochastic Processing Calculus. The contributions from this effort include exploring this new area of mathematics and in its potentially transformational application to each of the two research areas.

Broader Impact: This INSPIRE project is transformational in that it promises to deliver a new rigorous modeling and analytical framework that can encompass a broad range of emerging networking problems. New analytical and algorithmic results that will be developed for the general abstraction of stochastic processing network are expected to have broad applications in a diverse range of fields. More broadly speaking, the proposed work is transformational in that it will contribute to a larger body of 'Network Science'. Network Science is being recognized as an emerging field in its own right in that many of the mathematical foundations and network algorithmics developed in the networking community are finding wide applications in many other fields.

The design and analysis of network algorithmics, namely, techniques and principles behind the software and hardware systems running on high-speed Internet routers, has become a rich area of research. In general, network operators would like routers to deliver robust performance under a wide variety of, often unforeseen, operating conditions. To address this need, this project takes a first look into network algorithmics solutions that can guarantee a certain level of performance, not only under typical or average parameter settings as in prior studies, but also under all admissible parameter settings. Toward this goal, PIs propose to develop a novel mathematical approach, called worst-case large deviation theory that is needed to prove such universal lower bounds on performance.

This project consists of three closely connected research threads. First, the principal investigators (PIs) will develop solutions for distributed data streaming problems that can guarantee a certain level of performance, under all possible ways a given data set is partitioned into distributed subsets. Second, they will develop a rich family of load-balanced switching solutions that can guarantee high throughput and reasonably low delay under all admissible traffic workloads. Third, they will build mathematical connections between worst-case large deviation techniques they developed in the past several years for deriving such universal performance bounds in prior network algorithmics solutions, which they expect will shed light on the new mathematical problems they will encounter in the first two research threads.

This project will engage both graduate and undergraduate students through integrated classroom curriculum and research training that span multiple disciplines, from fundamental mathematics, algorithm design, to hardware implementation. The results will be broadly disseminated through publications, invited talks, tutorials, and open-sourcing of software developed for this project in accordance with the policies of each institution. The PIs will work closely with leading networking and systems solution providers to facilitate technology transfers. Further, both PIs are committed to outreach efforts at their corresponding campuses to broaden the participation of under-represented groups in research and higher education.

This project will contribute to the national need for well-educated scientists, mathematicians, engineers, and technicians by supporting the retention and graduation of high-achieving, low-income students with demonstrated financial need at the University of California San Diego, Imperial Valley College, and Southwestern College. University of California San Diego is a public research university and an emerging Hispanic Serving Institution (HSI). Imperial Valley College and Southwestern College are 2-year community colleges and fully recognized HSIs. Over its 5 year duration, this project will fund 370 student-years of full-time scholarships to approximately 185 unique students who are pursuing associate’s and bachelor’s degrees in engineering. Transfer-track students at their host institution will receive 2-year scholarships. This project focuses evidence-based high-impact practices on the transition experience endured by transfer students before, during, and after their engagements with 2-year and 4-year institutions. The major significance of this work, beyond financially supporting these students in need, will be implementing, assessing, and ultimately sharing a model educational framework grounded in Schlossberg’s Transition Theory. The transfer students, who undergo frequent and challenging transitions, will be supported with enrichment activities including mentorship, summer programming, research opportunities, and academic year technical and professional workshops.<br/><br/>The overall goal of this project is to increase Engineering degree completion of low-income, high-achieving undergraduates with demonstrated financial need. The challenges faced by transfer students on their journey toward a STEM degree must be better understood. Each transitional step, in, through, and out of their respective institutions requires well-directed support and guidance. This project aims to evaluate the diverse transitions experienced by transfer students and to intervene appropriately, using planned academic year and summer programming as vehicles to deliver the necessary support. Methodologies and results derived from this project will be shared through conference presentations and journal publications. This project is funded by NSF’s Scholarships in Science, Technology, Engineering, and Mathematics program, which seeks to increase the number of low-income academically talented students with demonstrated financial need who earn degrees in STEM fields. It also aims to improve the education of future STEM workers, and to generate knowledge about academic success, retention, transfer, graduation, and academic/career pathways of low-income students.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.