Heya! I’m a Ph.D. student at the University of Washington’s Paul G Allen School of Computer Science and Engineering advised by Professor Simon Peter. My current work revolves around analyzing data movement and mitigating its performance impact in contemporary applications [ISCA ‘24, ASPLOS ‘25]. I hope to explore this problem across the stack from the single-node architectural level to multi-machine distributed systems.

There are three main axes by which I’m doing this:
1) Avoiding (unnecessary) data movement, i.e., (MC)²,
2) Reducing (necessary) data movement, and
3) Hiding the overheads of data movement by overlapping it with other operations, i.e., POD-Attention.

Prior to this, I spent two years as a research assistant at the Indian Institute of Science, working with Professor Arkaprava Basu. Here, I worked on improving race detection in GPUs [ISCA ‘20, SOSP ‘21], and improving GPU performance through the use of NVM technology [ASPLOS ‘22, ‘23].

Publications

• [ASPLOS '25] POD-Attention: Unlocking Full Prefill-Decode Overlap for Faster LLM Inference

  Distinguished Artifact Award
  

  Aditya K Kamath, Ramya Prabhu, Jayashree Mohan, Simon Peter, Ramachandran Ramjee, Ashish Panwar
  ACM 30th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)

• [ISCA '24] (MC)²: Lazy MemCopy at the Memory Controller

  Aditya K Kamath, Simon Peter
  ACM/IEEE 51st Annual International Symposium on Computer Architecture (ISCA)

• [ISCA '24] Scalable, Programmable and Dense: The HammerBlade Open-Source RISC-V Manycore

  Dai C Jung, Max Ruttenberg, Paul Gao, Scott Davidson, Daniel Petrisko, Kangli Li, Aditya K Kamath et al.
  ACM/IEEE 51st Annual International Symposium on Computer Architecture (ISCA)

• [ASPLOS '23] Scoped Buffered Persistency Model for GPUs

  Shweta Pandey*, Aditya K Kamath*, and Arkaprava Basu
  (*Authors contributed equally to this work)
  ACM 28th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)

• [ASPLOS '22] GPM: Leveraging Persistent Memory from a GPU

  Shweta Pandey*, Aditya K Kamath*, and Arkaprava Basu
  (*Authors contributed equally to this work)
  ACM 27th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)

• [SOSP '21] iGUARD: In-GPU Advanced Race Detection

  Aditya K Kamath and Arkaprava Basu
  ACM SIGOPS 28th Symposium on Operating Systems Principles (SOSP)

• [ISCA '20] ScoRD: A Scoped Race Detector for GPUs

  Aditya K Kamath*, Alvin A George*, and Arkaprava Basu
  (*Authors contributed equally to this work)
  ACM/IEEE 47th Annual International Symposium on Computer Architecture (ISCA)

Still here? Curious to see what first contact with aliens might look like?
Check out this talk I gave at the ASPLOS 2023 “Wild and Crazy Ideas” session: