Research

Current Projects

 Selected Publications

 Professional Activities

Research Opportunity


The general research of our research group is in the area of design automation for electronic systems. More specifically, our current research interests include real time computing, embedded system design, power aware computing, reconfigurable computing, advanced computer architecture, and high performance computing. Currently, our research group is researching in the following projects.

 

For the past decade, we have experienced the tremendous impacts and benefits of service oriented computing. From Google maps, to Amazon or Ebay, service oriented computing is drastically changing our daily lives. To satisfy the large number of service requests, service providers not only need to build more data centers, to add more computing resources (e.g. CPU, memory, hard drive, and bandwidth, etc.) to the data centers, but also to develop techniques that can utilize the resource more effectively and efficiently to make it sustainable. While numerous efforts have been dedicated to reducing data centers’ energy consumption and carbon footprints, this project addresses another critical component, i.e. water footprint, that has been largely neglected. A large data center may consume millions of gallons of cooling water each day; in addition, data centers also indirectly consume an enormous amount of water via electricity usage (approximately 1.8 liters of water evaporation per kilowatt-hour electricity in the U.S.). Left unchecked, the growing water footprint of data centers can pose a severe threat to data center sustainability and eventually handicap availability of services in water-stressed areas. This project seeks to resource management techniques and analysis methods that can effectively utilize computing resources (water, power/energy, carbon footprint, etc) in datacenters and, at the same time, to provide quality-guaranteed services (end-to-end deadline meet ratios, throughput, etc.).

The continued advancements in semiconductor technology make it possible to integrate hundreds processing cores into one silicon die. However, as the transistor size continues to shrink, the increasing manufacturing defects have resulted in severe yield loss and made chip products’ profit to drop substantially. Process variations also cause the performance and power characteristics differ significantly from chip to chip. This non-determinism greatly exacerbates the complexity of developing, validating, and maintaining software built upon these chips, especially for mission-critical real-time embedded applications. We believe that the architectural virtualization is an effective strategy and will be the norm to address the extensive process variations and manufacturing defect problems on many-core platforms. In this collaboration effort, we seek to develop effective methods and techniques to virtualize hardware resources on many-core platform and thus isolate the underlying hardware non-determinisms without changing the operating system and application software.

This research seeks to address the temperature and power/energy consumption problem using real-time scheduling techniques, with a focus on the interplay between temperature and leakage power consumption. This project intends to study the system-level thermal models that can capture the temperature-leakage interdependency with high accuracy while remain simple enough for formal system level analysis, and also to develop and validate novel and effective real-time scheduling techniques under both single and multiple processor platforms.

The research goal of this project is to leverage current operating system (OS) functionality to support real-time embedded systems and meet the increasingly stringent power/energy constraints of mobile devices. Toward this goal, we emphasize a comprehensive approach drawn from state of the art in IC and computer architecture technology, real-time theory, and practical applications, with the primary research focus on the development of novel scheduling methods and decision functions. The principle of our approach is to exploit the power management capability in advanced IC and computer architecture technology, and balance the application requirements and power/energy consumption.

This project intends to develop a simulation platform, based on the Virtual Test Bench (VTB), that can simulate the cyber sub-system and physical sub-system in a coordinated and unified way. The primary purpose of current VTB is to simulate physical system with phenomenon crossing different disciplines. We intend to expand the scope of current VTB by developing appropriate cyber space models such as those for processors, memory/storage, software, network, and operating systems as well as building corresponding simulation kernel and other facilitates to enable the simulation of cyber and physical systems in an integrated environment.

The primary focus of this project is to research and develop appropriate reconfigurable platform that can be effectively used for high performance computing purpose, such as those in the hardware-in-the-loop simulation. Our current research includes FPGA design and synthesis, algorithm optimization, advanced computer architecture exploration, real-time communication routing in multi-core systems, etc.

Funding/Equipment Support


Our research is supported by the following government agency, company, and institutes. 

·