M
Minjang Kim
Researcher at Georgia Institute of Technology
Publications - 5
Citations - 162
Minjang Kim is an academic researcher from Georgia Institute of Technology. The author has contributed to research in topics: Speedup & Automatic parallelization. The author has an hindex of 5, co-authored 5 publications receiving 155 citations. Previous affiliations of Minjang Kim include Qualcomm.
Papers
More filters
Proceedings ArticleDOI
SD3: A Scalable Approach to Dynamic Data-Dependence Profiling
TL;DR: This paper proposes a scalable approach to data-dependence profiling that addresses both runtime and memory overhead in a single framework, called SD3, and reduces the runtime overhead by parallelizing the dependence profiling step itself and compress memory accesses that exhibit stride patterns and compute data dependences directly in a compressed format.
Proceedings ArticleDOI
Predicting Potential Speedup of Serial Code via Lightweight Profiling and Emulations with Memory Performance Model
TL;DR: Parallel Prophet projects potential parallel speedup from an annotated serial program before actual parallelization, which models many realistic features of parallel programs: unbalanced workload, multiple critical sections, nested and recursive parallelism, and specific thread schedulings and paradigms, which are hard to model in previous approaches.
Proceedings ArticleDOI
CHiP: A Profiler to Measure the Effect of Cache Contention on Scalability
TL;DR: This work presents an efficient method to identify such cases and determine whether a serial algorithm's use of shared memory caches will seriously impact its parallel execution and can help a programmer adjust their program's cache usage.
Dissertation
Dynamic program analysis algorithms to assist parallelization
Hyesoon Kim,Minjang Kim +1 more
TL;DR: This dissertation proposes Prospector, which consists of several new and enhanced program analysis algorithms that assist the actual parallelization steps: finding parallelization candidates, understanding the parallelizability and profits of the candidates, and writing parallel code.
Journal ArticleDOI
SD3: An Efficient Dynamic Data-Dependence Profiling Mechanism
TL;DR: This paper proposes an efficient approach to data-dependence profiling that can address both runtime and memory overhead in a single framework, called SD3, and reduces the runtime overhead by parallelizing the dependence profiling step itself and compress memory accesses that exhibit stride patterns and compute data dependences directly in a compressed format.