I currently lead the LaCASA lab (Laboratory for Advanced Computer Architectures and Systems at Alabama).
The LaCASA Laboratory seeks for advanced computer architectures and new techniques for designing and evaluation of computer systems.
In particular, the LaCASA Laboratory is currently focusing on the following:
- Secure Processors: [ICCD'07] [JEC'06]
[CASES'05]
[WASSA'04][ACMSE'04]
- Trace
Compression: [DCC'07] [TOMACS'07] [WWC'03]
[CAL'03]
- Bridging the CPU-Memory Speed Gap: [ACMSE'04] [SSST'03] [MICPRO'01][CONC'00]
- Performance
Evaluation: [SPE'04]
[ACMSE'04]
[SSST03] [WWWD'02]
- Workload Characterization
- TinyHMS
(Tiny Wireless Sensor Networks for Personal Health Monitoring):
[COMPCOMM’06] [JMM'06] [JNER'05] [SSST05b] [SSST05a]
Research impact: Our papers have been referenced in more than 200 journal and conference papers. We proposed a number of architectural techniques such as lazy prefetching, cache injection, runtime verification of instruction block signatures for software integrity and confidentiality, and stream-based trace compression. Some of these techniques are implemented in modern computer systems. We introduced cache injection and this work has been referenced by a number of U. S. patents (Click here to see the list). Our analysis of cache replacement policies has also been referenced by a number of U. S. patents (Click here to see the list). We also introduced a framework and developed microbenchmarks for the first successful reverse engineering of branch predictor units in modern high-end microprocessors (Click here for the microbenchmarks). We play with trace compression techniques (we invented and patented stream-based trace compression), and offer to the research community instruction and data address traces for SPEC CPU2K benchmarks.
In addition to our research in computer architecture we are excited about wireless sensor networks. We introduced and developed one of the first prototypes of a wireless body area network (WBAN) for personal health monitoring (Click here for details). Our work on WBANs includes all components of computer engineering, from sensor interfacing and hardware development to energy-efficient algorithms for signal processing, system software, application software, and communication protocols. The paper co-authored with Dr. Jovanov, "A Wireless Body Area Network of Intelligent Motion Sensors for Computer Assisted Physical Rehabilitation," is the most read paper in the Journal of NeuroEngineering and Rehabilitation of all time (Click here for the rankings). Our other paper on wireless body area networks "Wireless sensor networks for personal health monitoring: Issues and an implementation" is consistently ranked among Top 25 Hottest articles in Computer Communications journal.
