I am Mahyar Emami, a 5th year Ph.D. student at EPFL, supervised by James Larus.

I work at the intersection of hardware and software. I like to build customized compilers for customized hardware.

Previously I have been working on the StreamBlocks compiler. With StreamBlocks you can treat the notoriously difficult-to-program FPGAs like multi-core systems. StreamBlocks is capable of auto-partitioning computational tasks across hardware and software with profiling information so you don’t have to think too much about what should be accelerated in hardware and what should run on software. You can read about StreamBlocks in:

Mahyar Emami, Endri Bezati, Jörn W. Janneck, and James R. Larus. 2023. “Auto-Partitioning Heterogeneous Task-Parallel Programs with StreamBlocks“. In Proceedings of the International Conference on Parallel Architectures and Compilation Techniques (PACT ’22) — slides

I am currently working on an exciting brand-new RTL simulation accelerator, called Manticore. Manticore is a 225-core beast that can exploit the fine-grained parallelism of RTL workloads to seriously speedup up the digital design flow. You can read about Manticore in:

Mahyar Emami, Sahand Kashani, Keisuke Kamahori, Mohammad Sepehr Pourghannad, Ritik Raj, and James R. Larus. “Manticore: Hardware-Accelerated RTL Simulation with Static Bulk-Synchronous Parallelism.” arXiv e-prints (2023): arXiv-2301 (to appear in ASPLOS’24’).