Ashish Panwar

Publications

• vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention

  Ramya Prabhu, Ajay Nayak, Jayashree Mohan, Ramachandran Ramjee, Ashish Panwar

  Conditionally accepted to appear in ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) 2025.

  Preprint | Code

• Taming Throughput-Latency Trade-off in LLM Inference with Sarathi-Serve

  Amey Agrawal, Nitin Kedia, Ashish Panwar, Jayashree Mohan, Nipun Kwatra, Bhargav Gulavani, Alexey Tumanov, Ramachandran Ramjee

  Published in 18th USENIX Symposium on Operating Systems Design and Implementation (OSDI) 2024.

  Paper | Code

• VIDUR: A Large-Scale Simulation Framework for LLM Inference

  Amey Agrawal, Nitin Kedia, Jayashree Mohan, Ashish Panwar, Nipun Kwatra, Bhargav S. Gulavani, Ramachandran Ramjee, Alexey Tumanov

  Published in the 7th Annual Conference on Machine Learning and Systems (MLSys) 2024.

  Paper | Code | Demo

• SIGMA: Secure GPT Inference with Function Secret Sharing

  Kanav Gupta, Neha Jawalkar, Ananta Mukherjee, Nishanth Chandran, Divya Gupta, Ashish Panwar, Rahul Sharma

  Published in 24th Privacy Enhancing Technologies Symposium (PETS) 2024.

  Paper

• Address Scaling: Architectural Support for Fine-Grained Thread-Safe Metadata Management

  Deepanjali Mishra, Konstantinos Kanellopoulos, Ashish Panwar, Akshitha Sriraman, Vivek Seshadri, Onur Mutlu, Todd C Mowry

  Published in IEEE Computer Architecture Letters 2024.

  Paper

• SARATHI: Efficient LLM Inference by Piggybacking Decodes with Chunked Prefills

  Amey Agrawal, Ashish Panwar, Jayashree Mohan, Nipun Kwatra, Bhargav S. Gulavani, Ramachandran Ramjee

  Arxiv 2023.

  Preprint

• Trident: Harnessing Architectural Resources for All Page Sizes in x86 Processors

  Venkat Sri Sai Ram, Ashish Panwar, Arkaprava Basu

  Published in 54th IEEE/ACM International Symposium on Microarchitecture (MICRO) 2021.

  Paper | Code | Slides | Video

• nuKSM: NUMA-aware Memory De-duplication on Multi-socket Servers

  Akash Panda, Ashish Panwar, Arkaprava Basu

  Published in 30th International Conference on Parallel Architectures and Compilation Techniques (PACT) 2021.

  Paper | Code | Slides | Video

• Fast Local Page-Tables for Virtualized NUMA Servers with vMitosis

  Ashish Panwar, Reto Achermann, Arkaprava Basu, Abhishek Bhattacharjee, K. Gopinath, Jayneel Gandhi

  Published in 26th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) 2021.

  Paper | Code | Extended Abstract | Slides | Video

• Mitosis: Transparently Self-Replicating Page-Tables for Large-Memory Machines

  Reto Achermann, Ashish Panwar, Abhishek Bhattacharjee, Timothy Roscoe, Jayneel Gandhi

  Published in 25th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) 2020.

  Paper | Code | Slides | Video

• HawkEye: Efficient Fine-grained OS Support for Huge Pages

  Ashish Panwar, Sorav Bansal, K. Gopinath

  Published in 24th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) 2019.

  Paper | Code | Slides | Poster

• Making Huge Pages Actually Useful

  Ashish Panwar, Aravinda Prasad, K. Gopinath

  Published in 23rd ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) 2018.

  Paper | Code | Slides | Poster

• A Case for Protecting Huge Pages from the Kernel

  Ashish Panwar, Naman Patel, K. Gopinath

  Published in 7th ACM SIGOPS Asia-Pacific Workshop on Systems (APSys) 2016.

  Paper

• Towards Practical Page Placement for a Green Memory Manager

  Ashish Panwar and K. Gopinath

  Published in 22nd IEEE International Conference on High Performance Computing (HiPC) 2015.

  Paper | Slides