Sri Pramodh Rachuri

Stony Brook, New York    ยท   

I am a Ph.D. Candidate in the Department of Computer Science at Stony Brook University. I am a part of the PACE lab and am advised by Prof. Anshul Gandhi. I broadly work in the areas of systems and networks. My research spans across Sustainable ML Computing, Edge Computing, Distributed Systems, Near Data Processing, Computer Networks and Wireless Communication. I have previously served as a student representative on the department's graduate committee and as a member of the executive board of CS GSO.

Before joining SBU, I received a BTech (Hons) in Electrical Engineering from the Indian Institute of Technology (IIT) Bhilai in 2020. I wrote my thesis on the topic "Interstream Coding Based SD-WAN For Multiaccess Systems" under the guidance of Prof. Arzad Alam Kherani. I am also the recipient of the Director's Gold Medal in the batch of 2016-20. I am a founding executive member of the Alumni Association of IIT Bhilai and have served as a Vice President from 2020 to 2023. Currently, I am a Executive Committee member of the association. I am also a board member of PanIIT USA since 2023.

Experience

Research Intern

Nokia Bell Labs, New Jersey   |   Mentor: Martin Havemann, Dr. Novak Boskov

Analyzed the geographical and operational challenges of green power generation for shared GPU resources. Developed a surplus green energy-aware scheduling system and integrated it into a decentralized computation framework for collaborative edge nodes. The system leverages a blockchain-based ledger to verify the "greenness" of energy used for tasks.

Jun 2024 - Sep 2024

Research Intern

AT&T Labs, New Jersey   |   Mentor: Dr. Shankaranarayanan

Evaluated the requirements for Virtualized RAN (vRAN) pooling and discerned differences in radio resource consumption patterns between LTE macro and microcells, as well as NR (5G) macro and mmWave cells. Currently involved in estimating the benefits of resource pooling and employing statistical methods for placement decisions by using anonymized production data traces from real-world radio equipment deployments.

Jun 2022 - Aug 2022
(Ongoing collab)

Research Intern

IIT Kharagpur   |   Guide: Prof. Sudip Misra

sed the Nonstationary Multi-armed Bandit reinforcement learning method to design a solution for task distribution challenges in Fog computing. Demonstrated, through simulation, that the design scheme outperforms existing cloud and fog solutions, reducing overall latency by 59\%. Further details are available in the publication on D2CIT.

May 2019 - Jul 2019

Research Intern

IIT Hyderabad   |   Guide: Prof. Sri Rama Murty K.

Worked on implementing Deep Learning methods for speech separation using a single microphone (commonly referred to as the cocktail party problem) in TensorFlow. The approach involved designing a Deep Neural Network with multiple layers of BiLSTMs and using the Short-Time Fourier Transform (STFT) transformation of speech as input data.

May 2018 - Jul 2018

Publications

EcoEdgeInfer: Dynamically Optimizing Latency and Sustainability for Inference on Edge Devices

2024 ACM/IEEE Symposium on Edge Computing (SEC)

Sri Pramodh Rachuri, Nazeeer Shaik, Mehul Choksi, Anshul Gandhi

OVIDA: Orchestrator for Video analytics on Disaggregated Architecture

2024 ACM/IEEE Symposium on Edge Computing (SEC)

Manavjeet Singh, Sri Pramodh Rachuri, Bryan Bo Cao, Abhinav Sharma, Venkata Bhumireddy, Anshul Gandhi, Francesco Bronzino, Samir Das, Shubham Jain

Evaluating the energy impact of device parameters for DNN inference on edge

2023 International Green and Sustainable Computing Conference (IGSC)

Anurag Dutt*, Sri Pramodh Rachuri*, Ashley Lobo, Nazeer Shaik, Anshul Gandhi, Zhenhua Liu
*joint first-authorship

Optimizing Near-Data Processing for Spark

2022 IEEE International Conference on Distributed Computing Systems (ICDCS)

Sri Pramodh Rachuri, Arun Gantasala, Prajeeth Emanuel, Anshul Gandhi, Robert Foley, Peter Puhov, Theodoros Gkountouvas, Hui Lei

Traffic Splitting for Delay Jitter Control in Multi-access Systems

2022 Springer Telecommunication Systems (TELS)

Megha Sahu, Sri Pramodh Rachuri, Ahtisham Ali Ansari, Arzad A Kherani

Decentralized modular architecture for live video analytics at the edge

2021 ACM Workshop on Hot Topics in Video Analytics and Intelligent Edges (HotEdgeVideo)
(in conjunction with ACM MobiCom)

Sri Pramodh Rachuri, Francesco Bronzino, Shubham Jain

Multiarmed-bandit-based decentralized computation offloading in fog-enabled IoT

2021 IEEE Internet of Things Journal (IoTJ) - Vol 8 Issue 12

Sudip Misra, Sri Pramodh Rachuri, Pallav Kumar Deb, Anandarup Mukherjee

On limiting Delay and Jitter characteristics at application-layer of Multi-connected Systems

2020 IEEE 5G World Forum (5GWF)

Megha Sahu, Sri Pramodh Rachuri, Ahtisham Ali Ansari, Deepaknath Tandur, Arzad A Kherani

An SD-WAN Controller for Delay Jitter Minimization in Coded Multi-access Systems

2019 IEEE International Conference on Advanced Networks and Telecommunications Systems (ANTS)

Ahtisham Ali Ansari, Sri Pramodh Rachuri, Arzad A Kherani, Deepaknath Tandur

Network-coded sd-wan in multi-access systems for delay control

2019 International Conference on contemporary Computing and Informatics (IC3I)

Sri Pramodh Rachuri, Ahtisham Ali Ansari, Deepaknath Tandur, Arzad A Kherani, Sameer Chouksey