Year: 2018
Type: conference-paper
Abstract: We present Lerna, an end-to-end tool that automatically and transparently detects and extracts parallelism from data dependent sequential loops using speculation combined with a set of techniques including code profiling, dependency analysis, instrumentation, and adaptive execution. Speculation is needed to avoid conservative actions and detect actual conflicts. Lerna targets applications that are hard-to-parallelize due to data dependency. Our experimental study involves the parallelization of 13 applications with data dependencies. Results on a 24-core machine show an average of 2.7x speedup for micro-benchmarks and 2.5x for the macro-benchmarks.
Institutions: Alexandria University, Lehigh University, Virginia Tech
Language: English
Subfield: Hardware and Architecture
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
Award: G7515358678
OpenAlex ID: W2807864865
