Shagnik Pal
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Shagnik Pal has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Chester Rebeiro
Indian Institute of Technology Madras, Chennai, India
1 shared publication - Farhad Merchant
Bernoulli Institute and CogniGron, University of Groningen, Groningen, The Netherlands
1 shared publication - Jeeho Ryoo
Olsen College of Engineering and Science Fairleigh Dickinson University Vancouver
1 shared publication - Lizy Kurian John
The University of Texas at Austin
1 shared publication - Neeraja J. Yadwadkar
The University of Texas at Austin, Austin, USA
1 shared publication - Nikhilesh Singh
Indian Institute of Technology Madras
1 shared publication
Latest Publications
- Open Access
- 19 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 88-101
Oneiros: KV Cache Optimization through Parameter Remapping for Multi-tenant LLM Serving
- 19 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 88-101
Open AccessOneiros: KV Cache Optimization through Parameter Remapping for Multi-tenant LLM Serving
- 16 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Embedded Systems Letters
- Vol. 16 (4) , 433-436
ProMiSE: A Programmable Hardware Monitor for Secure Execution in Zero Trust Networks
With the inevitable adoption of Zero Trust Architectures (ZTA) for enterprise networks, there is a need to continuously gauge the security health of connected devices. This requires runtime monitoring of the devices in the network. The challenge, especially in resource-constrained environments, is to ensure trusted monitoring at a fine granularity. In this paper, we propose PROMISE, a framework that overcomes this challenge and provides an online non-tamperable metric called trust score to quantify the security health of devices in a ZTA network. We use real-time hardware tracking of micro-architectural signals in the CPU to compute the trust score in a security co-processor that is isolated from the devices computing stack. The trust score for each device is sent to the ZTA host for corresponding responses. We evaluate PROMISE on an open-source RISC-V processor with different threat vectors, including ransomware, return-oriented programming (RoP) attacks, and cache-based micro-architectural attacks. We also deploy the framework on an AMD Artix 7AC701 FPGA and present the area overheads.
- 16 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Embedded Systems Letters
- Vol. 16 (4) , 433-436
ProMiSE: A Programmable Hardware Monitor for Secure Execution in Zero Trust Networks
With the inevitable adoption of Zero Trust Architectures (ZTA) for enterprise networks, there is a need to continuously gauge the security health of connected devices. This requires runtime monitoring of the devices in the network. The challenge, especially in resource-constrained environments, is to ensure trusted monitoring at a fine granularity. In this paper, we propose PROMISE, a framework that overcomes this challenge and provides an online non-tamperable metric called trust score to quantify the security health of devices in a ZTA network. We use real-time hardware tracking of micro-architectural signals in the CPU to compute the trust score in a security co-processor that is isolated from the devices computing stack. The trust score for each device is sent to the ZTA host for corresponding responses. We evaluate PROMISE on an open-source RISC-V processor with different threat vectors, including ransomware, return-oriented programming (RoP) attacks, and cache-based micro-architectural attacks. We also deploy the framework on an AMD Artix 7AC701 FPGA and present the area overheads.