Education
Publications
Conference Papers
-
Splitwise: Efficient Generative LLM Inference Using Phase Splitting
Pratyush Patel, Esha Choukse, Chaojie Zhang, Íñigo Goiri, Aashaka Shah, Saeed Maleki, Ricardo Bianchini
ISCA 2024
Best Paper Nominee
-
Characterizing Power Management Opportunities for LLMs in the Cloud
Pratyush Patel, Esha Choukse, Chaojie Zhang, Íñigo Goiri, Brijesh Warrier, Nithish Mahalingam, Ricardo Bianchini
ASPLOS 2024
-
Hybrid Computing for Interactive Datacenter Applications
Pratyush Patel, Katie Lim, Kushal Jhunjhunwalla, Ashlie Martinez, Max Demoulin, Jacob Nelson, Irene Zhang, Thomas Anderson
Preprint
-
File Systems are not Enough: Rethinking the Storage API for Microsecond-Scale Cloud Applications
Ashlie Martinez, Katie Lim, Pratyush Patel, Irene Zhang, Dan Ports, Jacob Nelson, Thomas Anderson
Preprint
-
Srifty: Swift and Thrifty Distributed Neural Network Training on the Cloud
Liang Luo, Peter West, Pratyush Patel, Arvind Krishnamurthy, Luis Ceze
MLSys 2022
-
The Demikernel Datapath OS Architecture for Microsecond-scale Datacenter Systems
Irene Zhang, Amanda Raybuck, Pratyush Patel, Kirk Olynyk, Jacob Nelson, Omar Navarro Leija, Ashlie Martinez, Jing Liu, Anna Kornfeld Simpson, Sujay Jayakar, Pedro Henrique Penna, Max Demoulin, Piali Choudhury, Anirudh Badam
SOSP 2021
-
SoundWatch: Exploring Smartwatch-based Deep Learning Approaches to Support Sound Awareness for Deaf and Hard of Hearing Users
Dhruv Jain, Hung Ngo, Pratyush Patel, Steven Goodman, Leah Findlater, Jon Froehlich
ASSETS 2020
Best Artifact Award, Selected for CACM Research Highlights
-
The Virtual Block Interface: A Flexible Alternative to Conventional Virtual Memory Frameworks
Nastaran Hajinazar, Pratyush Patel, Minesh Patel, Konstantinos Kanellopoulos, Saugata Ghose, Rachata Ausavarungnirun, Geraldo Francisco de Oliveira Jr., Jonathan Appavoo, Vivek Seshadri, Onur Mutlu
ISCA 2020
-
Gandiva: Introspective Cluster Scheduling for Deep Learning
Wencong Xiao, Romil Bhardwaj, Ramachandran Ramjee, Muthian Sivathanu, Nipun Kwatra, Zhenhua Han, Pratyush Patel, Xuan Peng, Hanyu Zhao, Quanlu Zhang, Fan Yang, Lidong Zhou
OSDI 2018
-
Analytical Enhancements and Practical Insights for MPCP with Self-Suspensions
Pratyush Patel, Iljoo Baek, Hyoseung Kim, Raj Rajkumar
RTAS 2018
Best Presentation Award
-
A Server-Based Approach for Predictable GPU Access Control
Hyoseung Kim, Pratyush Patel, Shige Wang, Raj Rajkumar
RTCSA 2017
Best Paper Award
-
TimerShield: Protecting High-Priority Tasks from Low-Priority Timer Interference
Pratyush Patel, Manohar Vanga, Björn Brandenburg
RTAS 2017
Best Paper Award
Journal Papers
-
Towards Improved Power Management in Cloud GPUs
Pratyush Patel, Zibo Gong, Syeda Rizvi, Esha Choukse, Pulkit Misra, Thomas Anderson, Akshitha Sriraman
IEEE CAL 2023
-
SoundWatch: Deep Learning for Sound Accessibility on Smartwatches
Dhruv Jain, Hung Ngo, Pratyush Patel, Steven Goodman, Khoa Nguyen, Rachel Grossman-Kahn, Leah Findlater, Jon Froehlich
CACM Research Highlights 2022
-
A Server-Based Approach for Predictable GPU Access with Improved Analysis
Hyoseung Kim, Pratyush Patel, Shige Wang, Raj Rajkumar
Journal of Systems Architecture 2018
Workshop Papers and Posters
-
An Agile Pathway Towards Carbon-aware Clouds
Pratyush Patel, Theo Gregersen, Thomas Anderson
HotCarbon 2023
-
Designing Equitable Data Center Scheduling Systems
Sahana Rangarajan, Xuesi Chen, Pratyush Patel, Sara Mahdizadeh Shahri, Jaylen Wang, Akshitha Sriraman
CWIDCA at MICRO 2022
-
Extreme Memoization: Everything in a LUT!
Pratyush Patel, Luis Ceze
WACI at ASPLOS 2020
-
μTVM: Deep Learning on Bare-Metal Devices
Logan Weber, Pratyush Patel, Tianqi Chen
ARM Research Summit 2019
-
μTVM: Deep Learning on Bare-Metal Devices
Pratyush Patel, Tianqi Chen, Luis Ceze
TVM Conference 2018
Experience
Microsoft Research
Redmond, USA
June 2022 – September 2022
Research Intern, Cloud Efficiency Group
Mentors: Esha Choukse, Pulkit Misra
- Identified power reduction and overclocking opportunities on modern CPUs and GPUs across diverse datacenter workloads
- Designed and implemented a VM-aware power management framework prototype for heterogeneous servers
Microsoft Research
Bengaluru, India
September 2017 – May 2018
Research Intern, Systems Group
Mentors: Vivek Seshadri, Muthian Sivathanu
- Designed a hardware-based virtual memory interface and evaluated its impact on native and virtual machine performance
- Implemented deep-learning training job scheduling and language-level checkpointing in Gandiva
BITS Pilani
Goa, India
January 2016 – May 2016
Undergraduate Researcher
Mentor: Santonu Sarkar
- Developed a checkpoint toolkit that traces program call graphs and probabilistically places checkpoints
- Modified QEMU to trace assembly instructions executed by guest VMs to quantify instruction patterns
Infibeam
Ahmedabad, India
May 2015 – July 2015
Software Development Intern
- Containerized production servers for log management with Docker-based ELK stack
- Proposed design recommendations for porting an internal Java Struts web application to console mode
Teaching
Teaching Assistant, CSE 552 Distributed Systems, Spring 2024
Seminar Co-creator and Teaching Assistant, CSE 590X: The Tao of PhD, Winter 2022
Teaching Assistant, CSEP 548 Computer Architecture, Winter 2020
Mentor, Real-Time and Multimedia Systems Laboratory, Carnegie Mellon University, Fall 2017
Instructor, Operating Systems from Scratch, Technology Incubator Program, Fall 2015
Mentoring
Awards and Honours
Service
External Reviewer, OSDI 2024
Reviewer, CCPE Journal 2023
Proposed and co-created CSE 590X: The Tao of PhD, Winter 2022
UW CSE Diversity Committee, 2020-2022
Allen School Pre-Application Mentorship Service (PAMS) Mentor, 2020-2021
UW CSE Faculty Candidate DEI Reviews Committee, 2020-2021
UW CSE Prospective Students Committee, 2019, 2021
Graduate Student Mentor, 2019-2022
Artifact Evaluation Committee, ASPLOS 2020
Organizing Volunteer, CPSWeek 2017
External Reviewer, RTNS 2016
Secretary, ACM Student Chapter, BITS Goa, 2014–2016
Core Member, Mobile Applications Club, BITS Goa, 2014–2015