The HPC research group @ Tx State was formed in Fall 2007. The group conducts research on a variety of problems in the area of high-performance computing. The main research goal of the group is to develop adaptive software technology that can keep pace with the changing landscape of computer architecture and deliver portable high-performance on both multi-core and many-core systems.
Current Projects
- Shared-cache on Multi-core Systems
- Memory Hierarchy Optimizations for the GPU
- Autotuning for Scientific Applications
- Compiler-driven Superpage Allocation
- Power-aware Task Scheduling
News and Activity
12/2010
Suman Vara defends his thesis, Evaluating Shared-cache Performance with Microbenchmarks and Reuse Distance Analysis.
11/2010
Santosh presents poster at SC10
10/2010
Nvidia donates Tesla GPU to the HPC group
08/2010
Search space exploration paper selected as best paper at NPC 2010
06/2010
SRC to fund proposal on power-performance scheduling for multicore and manycore architectures.
06/2010
Santosh to spend summer as an intern at Freescale, Austin.
05/2010
Hammad Rashid defends his thesis, Parallel Knapsack Algorithms on Multicore Architectures. He will join VMWare in California.
05/2010
HPC groups papers accepted at iWAPT10, NPC10 and CSC10.
04/2010
Hammad presents poster at the Women in Science and Engineering (WISE) Conference held at Texas State University. The poster describes Hammad's work on optimizing dynamic programming-based algorithms for the knapsack problem.
04/2010
Apan presents paper on restructuring parallel loops at IPDPS10.
11/2009
Suman and Santosh present research posters at the International Research Conference at Texas State University
07/2009
IBM donates P6 machine to the HPC research group
03/2009
Paper on exploiting shared-caches on CMPs accepted at HPCC0902/2009
Superpage optimization paper accepted at ACMSE09
12/2008
Josh Magee defends his thesis, Automated Compiler Driven Superpage Allocation and its Applications
12/2008
Jason Cade defends his thesis, Balancing Data Locality and Parallelism for Improved Application Performance on Multi-core Platforms
12/2008
REP funds proposal on mapping OR algorithms onto multi-core systems
11/2008
Paper on model-guided tuning accepted by Journal of High-Performance Systems Architecture
09/2008
Dr. Qasem Awarded IBM Faculty Award for work on locality optimizations on shared-cache CMPs