Computer Architecture and Compilers

The architecture group at UCSD is at the cutting edge of innovation in processor and computer systems design.  Our current projects address the current "grand challenge" of computer architecture:  Finding power-efficient ways to find and exploit parallelism to continue scaling performance. We are addressing this challenge by exploring novel programming abstractions, CPU and storage system organization, execution strategies, micro-architectures, processor design methodologies, and circuit technologies.  Our group consists of 3 faculty and over 20 graduate students and is well-supported through generous funding from government and industrial sources.

Projects | People | Publications | Sponsors

Faculty

Steven Swanson
Dr. Steven Swanson
Michael Taylor
Dr. Michael Taylor
Dean Tullsen
Dr. Dean Tullsen

[More People]

Projects

ArsenalGreenDroid: Specialized Processors for Efficient Mobille Application Processors

We are developing a prototype processor called GreenDroid that is designed to dramatically reduce energy consumption in smart phones. GreenDroid will provide many specialized processors targeting key portions of the Google's Android smart phone platform. GreenDroid will reduce energy consumption for these codes by integrating Conservation Cores (C-cores).

Faculty: Steven Swanson and Michael Taylor

NVSLThe Non-volatile Systems Laboratory

The Non-volatile Systems Lab studies non-volatile, solid-state storage technologies and their applications at the architecture, system, and software levels.  We develop novel system and hardware architectures for advanced non-volatile memories, characterize existing technologies, and develop techniques to ensure the security of data stored in these devices.  It works with Center for Magnetic Recording Research, the San Diego Supercomputing Center, government agencies, and industrial partners to understand current and future non-volatile memory technologies and craft the system architectures, abstractions, and applications required to fully realize their potential.

Faculty: Steven Swanson

BandwidthSoftware Optimizations for Single Threads Running on CMPs

As the number of cores grows, Amdahl's Law increases the importance of serial execution performance. We are developing software techniques that allow us to accelerate a single thread using multiple, independent cores. Our results show that some simple techniques can significantly boost performance on using existing hardware. Read about it here.

Faculty: Dean Tullsen and Steven Swanson

nvslSD-VBS: The San Diego Vision Benchmark Suite

The San Diego Vision Benchmark Suite (SD-VBS) is a suite of diverse vision applications drawn from the vision domain. It is intended to help architects, compiler writers, and system designers study the construction of future systems that excel at vision-oriented applications. Additionally, vision codes tend to have a fair amount of parallelism, which makes them a good candidate for formulating future multicore and parallel architectures.

Faculty: Michael Taylor

Pyrprof: A Multicore Software Engineering Profiling Tool

Pyrprof is a parallel programming tool intended to help a programmer quickly discover parallelism in their programs. It outputs a list of code regions that is ordered by the potential speedup obtained from parallelization. Pyrprof provides a drop-in replacement for gcc and operates similarly to gprof.

Faculty: Michael Taylor

People

PhD Students

MS Students

  • Ameen Akel
  • Joe Auricchio
  • Slavik Bryksin
  • Jose Lugo-Martinez
  • Scott Ricketts
  • Daniel Stufflebean
  • Hyojin Sung

 

Sponsors

National Science Foundation (NSF) Intel DARPA HP STMicro Samsung Western Digital