معرفی
Akshay Satish Bhosale is a Postdoctoral Research Associate at the University of Cambridge's Department of Computer Science and Technology, working within the Computer Architecture group under the supervision of Dr. Tim Jones. His research focuses on developing advanced compiler techniques for exploiting intra-core parallelism in processors, with particular emphasis on optimization techniques and intermediate representations for efficient code generation.
His educational background includes a PhD from the University of Delaware under Dr. Rudi Eigenmann and a Bachelor's degree in Electronics Engineering from the University of Mumbai, India (2016).
Dr. Bhosale's research interests span multiple areas of compiler technology and parallel computing:
- Compiler Optimization techniques for modern processor architectures
- Hardware-Software co-design methodologies
- Automatic Parallelization of complex computational patterns
- High Performance Computing applications
- Intermediate representation development
His publication record demonstrates a consistent focus on solving the challenging problem of automatic parallelization of subscripted subscript patterns in scientific applications. Through his work, he has developed novel compile-time algorithms based on symbolic range aggregation that can detect properties of subscript arrays to prove parallelism in loops that conventional techniques cannot handle. His research has been published in top-tier conferences including PPoPP, ICS, and IPDPSW.
Dr. Bhosale is actively involved in several significant research projects:
- ParaSol project (funded by EPSRC UK) - developing compiler analyses and transformations to extract parallelism for future processors
- Software Prefetching implementation in LLVM infrastructure
- Automatic Parallelization of Irregular Computations using the Cetus source-to-source translator
He maintains an active presence in the compiler research community through his GitHub repositories and contributions to open-source compiler infrastructure projects, particularly the Cetus compiler framework for automatic parallelization.
