Research Summary

Checkout my Google Scholar profile for more upto date list of publications.

Thermal Modeling and Management of Multi-Chiplet Architectures

Emerging 2.5D and 3D multi-chiplet architectures present unique thermal challenges due to their dense integration and high compute density. My work in this area focuses on developing multi-fidelity thermal modeling frameworks, MFIT, which blend accurate and efficient methods which intern can be used to manage thermal hotspots, optimize chiplet placement, and explore design trade-offs. These models enable faster simulations and effective runtime management, ensuring thermal stability in advanced systems.

This research addresses critical issues in heterogeneous integration, such as inter-chiplet heat dissipation and crosstalk, contributing to reliable and efficient next-generation computing platforms.

Published Papers

1. Lukas Pfromm*, Alish Kanani*, Harsh Sharma, Parth Solanki, Eric Tervo, Jaehyun Park, Janardhan Rao Doppa, Partha Pratim Pande, and Umit Y. Ogras, “MFIT: Multi-Fidelity Thermal Modeling for 2.5 D and 3D Multi-Chiplet Architectures” arXiv preprint (Under Review) 2024. Paper, Blog, Code
2. Jaehyun Park, Alish Kanani, Lukas Pfromm, Harsh Sharma, Parth Solanki, Eric Tervo, Janardhan Rao Doppa, Partha Pratim Pande, and Umit Y. Ogras, “Thermal Modeling and Management Challenges in Heterogenous Integration: 2.5D Chiplet Platforms and Beyond” - VTS, 2024. Paper

Runtime Optimization using Machine Learning

Dynamic runtime optimization focuses on balancing execution time, energy, temperature, and resource utilization in mordern computing systems, including heterogeneous architectures and Domain-Specific Systems-on-Chip (DSSoCs). My work in this area develops ML-based scheduling algorithms that adapt to runtime changes, ensuring efficient allocation across general-purpose cores and specialized accelerators. These frameworks aim to improve system performance and energy efficiency, addressing challenges like dynamic workloads and resource contention.

This is an exciting area with ongoing developments—stay tuned for more!

Published Papers

1. Alper A. Goksoy, Alish Kanani, Satrajit Chatterjee, and Umit Y. Ogras, “Runtime Monitoring of ML-Based Scheduling Algorithms Toward Robust Domain-Specific SoCs” ESWEEK- TCAD, 2024. Paper, Blog

Hardware Accelerators for ML models

As part of my research on hardware accelerators for machine learning, I worked on LightFPGA, a project focused on the scalable and automated FPGA acceleration of LightGBM, a popular gradient boosting framework for machine learning applications.

I am currently working on developing more advanced accelerators tailored for modern machine learning models, including generative models, to meet the growing demand for high-performance, resource-efficient solutions in the AI landscape. Stay tuned for more updates!

Published Papers

1. Alish Kanani*, Swar Vaidya* and Harshit Agarwal, “LightFPGA: Scalable and Automated FPGA Acceleration of LightGBM for Machine Learning Applications” VDAT, 2021. Paper, Slides, Code

Approximate Computing and Applications

Approximate computing trades accuracy for efficiency, making it a perfect fit for energy-constrained systems and error-resilient applications.

I worked on several exciting projects in this area, such as developing ReARM, a reconfigurable approximate multiplier, and ACA-CSU, a carry-selection-based adder, to optimize power and performance for tasks like image processing. My work extended to ApproxBioWear, targeting efficient arithmetic for wearable biomedical devices, and even explored the feasibility of approximation in communication systems. Most recently, Ellora applied these concepts to create low-power radar processors, blending approximate computing with advanced OFDM techniques.

Together, these projects showcase a broad application of approximation, pushing the boundaries of energy efficiency across diverse domains.

Published Papers

1. Rajat Bhattacharjya, Alish Kanani, A Anil Kumar, Manoj Nambiar, M Girish Chandra, Rekha Singhal, “Ellora: Exploring Low-Power OFDM-based Radar Processors using Approximate Computing” – LASCAS 2024. Paper
2. Ish Kool, Alish Kanani, Rajat Bhattacharjya, “Approximating Communication Systems: Reality or Fantasy?” – HiPC 2021. Poster
3. Alish Kanani, Rajat Bhattacharjya and Dip Sankar Banerjee, “ApproxBioWear: Approximating Additions for Efficient Biomedical Wearable Computing at the Edge” – EMBC 2021. Paper, Slides
4. Alish Kanani, Jigar Mehta and Neeraj Goel, “ACA-CSU: A Carry Selection Based Accuracy Configurable Approximate Adder Design” – ISVLSI 2020. Paper, Slides, Code
5. Rajat Bhattacharjya, Alish Kanani and Neeraj Goel, “ReARM: A Reconfigurable Approximate Rounding-Based Multiplier for Image Processing” – VDAT 2020. Paper, Slides

* Equal contributions