IndoSys 2026

About IndoSys

Indian Symposium on Computer Systems

Computer systems research is critically important in the post-Moore era. Today's systems must support diverse big data platforms, address the scalability challenges of deep learning models, and efficiently exploit heterogeneous accelerators, including emerging quantum hardware, deployed across both cloud and edge environments.

The Indian Symposium on Computer Systems (IndoSys) was conceived as a forum to strengthen and grow the computer systems research ecosystem in India. IndoSys 2026 is the 8th edition of the symposium. The series began in 2014, with a brief hiatus during the COVID-19 period. Compared to several other research domains, the number of academic and industrial research labs in India focused on systems is relatively small. IndoSys aims to help address this gap by providing a platform for students, faculty, researchers, and industry practitioners working in (or interested in) computer systems.

The need for systems research has become even more pressing due to the data deluge and the widespread adoption of billion-parameter deep learning models across application domains. IndoSys will serve as a venue where researchers can present preliminary work, receive constructive feedback, and refine their ideas. A key distinguishing feature of systems research (relative to theory-focused work) is rigorous empirical evaluation; careful benchmarking and thorough experimentation are central to producing convincing systems results.

Details on previous editions on IndoSys are here.

The Heterogeneous Compute Continuum: Abstractions and Platforms for Edge AI, Agentic Clouds and Quantum Workflows

Yogesh Simmhan, Professor, Department of Computational and Data Sciences, IISc, Bangalore

Abstract

The compute landscape is broadening into a heterogeneous continuum — from power-constrained edge accelerators to elastic cloud platforms and nascent quantum backends — even as AI-driven cyber-physical applications such as drone fleets and urban mobility grow more demanding. This talk presents a systems-centric agenda spanning three frontiers.

At the edge, we examine running modern AI on constrained hardware: the performance, throughput, and power trade-offs of LLM and DNN inference on Jetson Orin-class accelerators, adaptive edge–cloud scheduling for personalized UAV fleets, and resilient federated learning over heterogeneous devices, illustrated through drone-based sensing and urban mobility verticals.

Moving to the cloud, we use serverless computing (FaaS) as a unifying abstraction for two emerging workload classes. First, we show how FaaS platforms can be re-architected to support MCP-enabled agentic workflows, informed by empirical characterization of FaaS behavior on public clouds. Second, we extend the same orchestration lens to quantum, presenting the choreography and profiling of quantum–classical FaaS workflows on hybrid clouds, where reusable patterns yield substantial gains in time and cost while coping with uncertain QPU queues.

We close with open challenges in building agile and efficient platforms that span the edge-to-quantum continuum.

Speaker Bio

Yogesh Simmhan is a Professor in the Department of Computational and Data Sciences at the Indian Institute of Science, Bangalore. His research explores scalable software platforms, algorithms, and applications on distributed systems, spanning Cloud and Edge Computing, Temporal Graph Processing, and Scalable Machine Learning for Big Data and IoT applications. He has published over 100 peer-reviewed papers and received numerous awards including the Best Paper Award at IEEE CLOUD 2019, the Distinguished Paper Award at EuroPar 2018, and the Swarna Jayanti Fellowship in 2019. He received the IEEE TCSC Award for Excellence in Scalable Computing (Mid Career Researcher) in 2020.

Yogesh holds a Ph.D. in Computer Science from Indiana University, Bloomington, and was previously a Research Assistant Professor at the University of Southern California and a Postdoc at Microsoft Research, San Francisco. He is a Distinguished Member of ACM, a Distinguished Contributor of the IEEE Computer Society, and serves on the ACM India Executive Council.

Work-in-Progress Papers

• LIOS: Runtime Precondition Checking for LLM Inference Optimizations
  Vaibhav Bajaj (IIT Delhi); Subharshi Roy (IIT Delhi)
• Trace-Adaptive Latency Prediction on GPUs for Deep Learning Models with Unseen Operators
  Parikshit Gehlaut (IIT Dharwad); Nikhil Hegde (IIT Dharwad)
• Towards Resilient Autonomous Platooning: A Self-Healing Framework via Multi-Agent Reinforcement Learning
  Arunkumar Arulappan (VIT University, Vellore)
• Ternary Spectrum Sensing for Cognitive Radio using PA-HNet
  Anjani Gupta (IIT Ropar); Abhinandan Prasad (IIT Palakkad)
• From Logical to Physical Schedules: Evaluating Ray on a Distributed LCS Workload
  Amish Kansal (IIT Delhi)
• Towards Optimizing Vision Transformers
  Nithin Damarla (IIT Dharwad); Nikhil Hegde (IIT Dharwad)
• High-Performance Matrix Multiplication on the Hexagon DSP Using SIMD and Multithreading
  Daram Vishnuvardhan (IIT Hyderabad); Shanmuga Vignesh (IIT Hyderabad); Sathya Peri (IIT Hyderabad)
• Near-Memory Tensor Mapping for Real-Time 6G AI-RAN Pre-Inference Conditioning
  Khyati Sharma (Indira Gandhi Delhi Technical University for Women); Apoorva Tripathi (Indira Gandhi Delhi Technical University for Women)
• Digital Twin Framework for Validating AI Training Networks in Clos Data Center Fabrics
  Suresh V J (VIT Vellore); Arunkumar Arulappan (VIT Vellore)
• A Full-stack Simulator for Heterogeneous AI Training Clusters
  Sumit Kumar (IIIT Delhi); Sayantan Dasgupta (IIIT Delhi); Kushal Mitra (IIIT Delhi); Meet Dadhania (IIT Hyderabad); Rohan Sudhir Basugade (IIIT Delhi); Praveen Tammana (IIT Hyderabad); Satananda Burla (Marvell Technology Inc.); Abed Mohammad Kamaluddin (Marvell Technology Inc.); Rinku Shah (IIIT Delhi)

Invited Papers

• Designing Size-aware Cryptographic Primitives for FPGA-based Accelerators
  Aditya Peer (IIIT Delhi); Rohan Sudhir Basugade (IIIT Delhi); Siddhant (IIIT Delhi); Neeraj Kumar Yadav (IIIT Delhi); Agamdeep Singh (IIIT Delhi); Praveen Tammana (IIT Hyderabad); Sumit Darak (IIIT Delhi); Rinku Shah (IIIT Delhi)
• LoRaHART: Hardware-Aware Real-Time Scheduling for LoRa
  Niraj Kumar (IIT Goa); Soumya Ranjan Sahoo (NTU Singapore); Amalinda Gamage (NUS Singapore); Arvind Easwaran (NTU Singapore)
• Lock Shielding: A General Technique for Misuse-Resilient Locks
  Vivek Shahare (IIT Dharwad); Nikhil Hegde (IIT Dharwad); Milind Chabbi (Uber Technologies Inc.)
• Cross-Platform, Cross-Framework Development of Hybrid-Parallel Matrix-Multiplication Codes
  Vyuhita Bonthu (IIT Dharwad); Nikhil Hegde (IIT Dharwad)
• Platooning Without Leaders: A Performance-Driven Reframing of Cooperative Vehicle Systems
  Arunkumar Arulappan (VIT University, Vellore)
• RA Scoreboarding: Return Address Stack Protection in RISC-V
  Lalit Sharma (IIT Ropar); Satya Jaswanth Badri (BITS Pilani); Neeraj Goel (IIT Ropar)
• VIKRAM: Virtual-address Informed K-level Radix Acceleration Mechanism
  Pravesh Jamgade (IIT Ropar)
• Strata: Proactive Page Placement in Hybrid Memory Systems
  Upasna (IIT Ropar); Venkata Kalyan Tavva (IIT Ropar)
• RADP: Region-Aware Data Placement for Write Amplification Reduction in SSDs
  Aditya Rathi (IIT Ropar); Waqar Hassan Mir (IIT Ropar); Venkata Kalyan Tavva (IIT Ropar); Neeraj Goel (IIT Ropar)

Authors are invited to submit research manuscripts that demonstrate current research in all areas of computer systems such as Operating Systems, Programming Languages, Computer Networks, Compilers, Distributed Systems, High Performance Computing, Cloud Computing, Mobile Computing, Database, Compilers, Computer Architecture, Software-assisted Networking, Network Security, Blockchain systems, Systems for ML/DL etc. The papers may be submitted in one of the following tracks:

Invited Papers

We invite previously published research works that appeared in reputable conferences/journals in the above areas. (1-page Extended Abstract). Accepted invited papers will present a "gong show" teaser, followed by a poster presentation, enabling broader exposure and facilitating follow-up discussions with experts.

Work-in-progress papers

We invite unpublished manuscripts describing ongoing research on systems problems. (Maximum 4 pages). Accepted WIP papers will be allocated a 15–20 minute talk slot. The goal is to foster in-depth discussion, mentoring and actionable feedback to help authors strengthen ongoing work and publish the full work at leading forums.

Awards

IndoSys 2026 will recognize outstanding contributions with a Best Poster Award and a Best Presentation Award.

• Tutorial-1 : Simulating AI Infrastructure at Scale
• Tutorial-2 : Fault Tolerance in Distributed Stream Processing Engines

The tutorial sessions are an important part of the Systems ecosystem building exercise. These sessions will be structured around specific themes/topics and will focus on key concepts, hands-on skills and exercises. The aim is to provide sufficient background and exposure to tools and software for participants to build the preliminary skill set and confidence to subsequently explore topics on their own and/or with some guidance. Based on the number of sessions, they may be scheduled in parallel during the event.

Themes and topics in the following areas (non-exhaustive) are of interest: Operating Systems, Programming Languages, Computer Networks, Compilers, Distributed Systems, High Performance Computing, Cloud Computing, Mobile Computing, Software-assisted Networking, Network Security, Blockchain systems, Systems for ML, etc.

This is a call for expression of interest to run a tutorial at IndoSys 2026. Each topic/theme of the tutorial will have sessions on two days of at least 1.5 hours each, for a total of 3 hours. You can expect participants to bring their own laptops for these sessions, have Internet connectivity and can be seated in lab-like classrooms. Please specify additional requirements, if any, as part of the proposal details, and we will try to accommodate the same.

Proposal Details

Following are the details to submit your proposal:

• Title of tutorial
• List of topics, specifically which of them will be hands on
• Pre-requisites, background of participants
• Expected outcomes
• Reference material
• Additional setup/equipment requirements (lab space, network, software packages, servers, …)

Please email your submissions as a PDF file to rinku@iiitd.ac.in with the subject "CFT: IndoSys 2026 <Tutorial Title>" by June 25, 2026.

Local accommodation and need-based travel support for 1-2 student TAs is available.

Author Manuscript and Submission Guidelines

Submitted manuscripts should be structured as technical papers and may not exceed four (4) single-spaced double-column pages using 10-point size font on 8.5×11 inch pages (ACM conference style), including figures, tables and 1 page for references. The IEEE conference style templates for MS Word and LaTeX provided by IEEE eXpress Conference Publishing are available for download. See the latest versions here. Electronic submissions must be in the form of a readable PDF file. All submitted manuscripts will be reviewed by the Program Committee.

Papers submitted to track 2 must represent original unpublished research that is not currently under review for any other conference or journal. These can be ongoing work that may not be ready for full submission. Authors may contact the Steering Committee Members and the Program Chairs for further information or clarification. Note that the goal of IndoSys is to provide a platform for feedback, discussion and mentorship on research work and writeup. The submissions will not be indexed/published and will not have a DOI.

How to Submit Papers

• Go to the link: CMT https://cmt3.research.microsoft.com/Indosys2026.
• Sign in or create an account. If you don’t have an account, click "Register" to create one with your email.
• Create a submission. Under the "Submissions" menu, you will see a "Create new submission" button. Select the appropriate track (WIP Papers, Invited Papers). Add details for your paper such as title, abstract, authors, affiliation, PDF of paper, etc.
• Fill in the Additonal Questions section. Provide additional details requested in the submission form relevant to each track, such as institution, experience, etc.
• Confirm and submit

Submission Review Process

All submitted manuscripts will be reviewed by the Program Committee members and undergo a rigorous review process. Submissions will be judged by at least two PC members on correctness, originality, technical strength, significance, potential impact, quality of presentation, and interest and relevance to the conference scope. The decision for the submission may be Accept or Reject. Papers that are accepted must be revised as per the reviewers’ suggestions.

A good-quality submission should articulate its motivation, key contributions, limitations of prior work, methodology and initial experimental results if any. Selected papers will be asked to present a short talk and/or a poster. A limited number of travel grants may be available for one author of selected papers.

Use of AI-assisted text

You may use tools such as ChatGPT, Grammarly, or other AI assistants to help improve the submission text quality. We recommend that you use these services only for language/grammar correction. The use of any AI-generated text must be disclosed in the acknowledgments section. The sections of the paper that use AI-generated text shall have a citation to the AI system used to generate the text.

All deadlines are 11:59PM Indian Standard Time (IST, UTC+0530).

• Papers submissions due: June 3, 2026 June 10, 2026 June 12, 2026
• Author Notification: June 24, 2026
• Symposium: July 20-21, 2026

Registration for IndoSys 2026 is now open. Please use the link below to register:

Register here: https://payments.iiit.ac.in/indosys2026/registration.php

Early Bird Registration Deadline: July 12, 2026

Day 1 — Monday, 20 July 2026

Day 2 — Tuesday, 21 July 2026

Invited Faculty Talks

• Abhinandan S P, IIT Palakkad
• Ajay Singh, IIT Gandhinagar
• Niraj Kumar, IIT Goa
• Ponnurangam Kumaraguru (PK), IIIT Hyderabad
• Rajshekar K, IIT Dharwad
• Sajal Das, Missouri S&T
• Sanchari Sen, IISc Bangalore
• Sathya Peri, IIT Hyderabad
• Sumit Kumar Mandal, IISc Bangalore
• Vishwesh Jatala, IIT Bhilai

Program Committee

• Abhilash Jindal (IIT Delhi)
• Abhinandan S Prasad (IIT Palakkad)
• Nitin Auluck (IIT Ropar)
• Padmanabha V Seshadri (IBM Research)
• Priyanka Naik (IBM Research, India)
• Raghavendra Kanakagiri (IIT Tirupati)
• Rajshekar K (IIT Dharwad)
• Sandeep Chakraborty (IIT Kharagpur)
• Sathya Peri (IIT Hyderabad)
• Urbi Chatterjee (IIT Kanpur)
• Varsha Apte (IIT Bombay)
• Vivek Kumar (IIIT Delhi)

Organization Committee

Steering Committee

• Kishore Kothapalli (IIIT, Hyderabad)
• Purushottam Kulkarni (IIT, Bombay)
• Yogesh Simmhan (IISc, Bangalore)

General Chairs

• Dip Sankar Banerjee (IIT, Jodhpur)
• Rinku Shah (IIIT, Delhi)

Program Chairs

• Nikhil Hegde (IIT, Dharwad)
• Preeti Malakar (IIT, Kanpur)

The symposium acknowledges the financial and in-kind support given by various organizations:

• The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.

Sponsors

Marvell Technology

Anusandhan National Research Foundation (ANRF)

DST PURSE — Promotion of University Research and Scientific Excellence

Institute Innovation Council, IIITD

Collaborating Institutions

Indraprastha Institute of Information Technology Delhi (IIIT-Delhi)

International Institute of Information Technology, Hyderabad

Indian Institute of Technology, Kanpur

Indian Institute of Science, Bangalore

Indian Institute of Technology, Bombay

Indian Institute of Technology, Jodhpur

Indian Institute of Technology, Dharwad