Alish Kanani
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Alish Kanani has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Umit Yusuf Ogras
University of Wisconsin - Madison, Madison, WI, USA
8 shared publications - Harsh Sharma
Washington State University
4 shared publications - Janardhan Rao Doppa
Washington State University
4 shared publications - Partha Pratim Pande
Washington State University
4 shared publications - Jaehyun Park
University of Ulsan, Ulsan, Ulsan, Republic of Korea
3 shared publications - Lukas Pfromm
Electrical and Computer Engineering, University of Wisconsin-Madison
3 shared publications
Latest Publications
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
LEXI: Lossless Exponent Coding for Efficient Inter-Chiplet Communication in Hybrid LLMs
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessLEXI: Lossless Exponent Coding for Efficient Inter-Chiplet Communication in Hybrid LLMs
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
DUET: Disaggregated Hybrid Mamba-Transformer LLMs with Prefill and Decode-Specific Packages
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessDUET: Disaggregated Hybrid Mamba-Transformer LLMs with Prefill and Decode-Specific Packages
- Top Cited PapersOpen Access
- 10 November 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
- Vol. 31 (1) , 1-27
MFIT : Multi-FIdelity Thermal Modeling for 2.5D and 3D Multi-Chiplet Architectures
Rapidly evolving artificial intelligence and machine learning applications require ever-increasing computational capabilities, while monolithic 2D design technologies approach their limits. 2.5D/3D heterogeneous integration of smaller chiplets using advanced packaging has emerged as a promising paradigm for addressing this limit and meeting performance demands. These approaches offer a significant cost reduction and higher manufacturing yield than monolithic 2D integrated circuits. However, the compact arrangement and high compute density of these systems exacerbate thermal management challenges, potentially compromising performance. Addressing these thermal modeling challenges is critical, especially as system sizes grow and different design stages require varying levels of accuracy and speed. Since no single thermal modeling technique meets all these needs, this article introduces MFIT, a range of multi-fidelity thermal models that effectively balance accuracy and speed. These multi-fidelity models can enable efficient design space exploration and runtime thermal management. Our extensive testing on systems with 16, 36, and 64 2.5D integrated chiplets and 16×3 3D integrated chiplets demonstrates that these models can reduce execution times from days to mere seconds and milliseconds with negligible loss in accuracy.
- 10 November 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
- Vol. 31 (1) , 1-27
Top Cited PapersOpen AccessMFIT : Multi-FIdelity Thermal Modeling for 2.5D and 3D Multi-Chiplet Architectures
Rapidly evolving artificial intelligence and machine learning applications require ever-increasing computational capabilities, while monolithic 2D design technologies approach their limits. 2.5D/3D heterogeneous integration of smaller chiplets using advanced packaging has emerged as a promising paradigm for addressing this limit and meeting performance demands. These approaches offer a significant cost reduction and higher manufacturing yield than monolithic 2D integrated circuits. However, the compact arrangement and high compute density of these systems exacerbate thermal management challenges, potentially compromising performance. Addressing these thermal modeling challenges is critical, especially as system sizes grow and different design stages require varying levels of accuracy and speed. Since no single thermal modeling technique meets all these needs, this article introduces MFIT, a range of multi-fidelity thermal models that effectively balance accuracy and speed. These multi-fidelity models can enable efficient design space exploration and runtime thermal management. Our extensive testing on systems with 16, 36, and 64 2.5D integrated chiplets and 16×3 3D integrated chiplets demonstrates that these models can reduce execution times from days to mere seconds and milliseconds with negligible loss in accuracy.
- Open Access
- 27 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Solid-State Circuits Society
- Vol. 5, 410-423
CHIPSIM: A Co-Simulation Framework for Deep Learning on Chiplet-Based Systems
Due to reduced manufacturing yields, traditional monolithic chips cannot keep up with the compute, memory, and communication demands of data-intensive applications, such as rapidly growing deep neural network (DNN) models. Chipletbased architectures offer a cost-effective and scalable solution by integrating smaller chiplets via a network-on-interposer (NoI). Fast and accurate simulation approaches are critical to unlocking this potential, but existing methods lack the required accuracy, speed, and flexibility. To address this need, this work presents CHIPSIM, a comprehensive co-simulation framework designed for parallel DNN execution on chiplet-based systems. CHIPSIM concurrently models computation and communication, accurately capturing network contention and pipelining effects that conventional simulators overlook. Furthermore, it profiles the chiplet and NoI power consumptions at microsecond granularity for precise transient thermal analysis. Extensive evaluations with homogeneous/heterogeneous chiplets and different NoI architectures demonstrate the frameworks versatility, up to 340 accuracy improvement, and power/thermal analysis capability.
- 27 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of the Solid-State Circuits Society
- Vol. 5, 410-423
Open AccessCHIPSIM: A Co-Simulation Framework for Deep Learning on Chiplet-Based Systems
Due to reduced manufacturing yields, traditional monolithic chips cannot keep up with the compute, memory, and communication demands of data-intensive applications, such as rapidly growing deep neural network (DNN) models. Chipletbased architectures offer a cost-effective and scalable solution by integrating smaller chiplets via a network-on-interposer (NoI). Fast and accurate simulation approaches are critical to unlocking this potential, but existing methods lack the required accuracy, speed, and flexibility. To address this need, this work presents CHIPSIM, a comprehensive co-simulation framework designed for parallel DNN execution on chiplet-based systems. CHIPSIM concurrently models computation and communication, accurately capturing network contention and pipelining effects that conventional simulators overlook. Furthermore, it profiles the chiplet and NoI power consumptions at microsecond granularity for precise transient thermal analysis. Extensive evaluations with homogeneous/heterogeneous chiplets and different NoI architectures demonstrate the frameworks versatility, up to 340 accuracy improvement, and power/thermal analysis capability.
- 15 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 45 (6) , 2840-2855
HeMu: Energy-Efficient DNN Inferencing via Heterogeneous Multichiplet Architectures
Processing-in-memory (PIM) is a promising solution to accelerate deep neural networks’ (DNNs) inference. However, the area and fabrication costs of monolithic PIM accelerators increase with growing complexity of DNN models. Hence, the design of PIM-based DNN accelerators should leverage the benefits introduced by the emerging chiplet-based 2.5-D architectures. Moreover, chiplets can be designed with various PIM devices, including resistive-random access-memory (ReRAM), ferroelectric field-effect transistor (FeFET), phase change memory (PCM), MRAM, and static random access-memory (SRAM). Each option offers advantages and limitations in terms of power, latency, area, and reliability. None of these PIM-based chiplets has all the desirable features necessary to design high-performance, energy-efficient, and reliable DNN accelerators. Hence, a thorough design space exploration (DSE) is needed to integrate different types of PIM-based chiplets in a single 2.5-D platform. This article proposes a design and optimization framework called HeMu to enable the design of high-performance and energy-efficient multichiplet architectures for accelerating DNN inferencing. HeMu performs a DSE to combine the benefits of more than one type of PIM-based chiplet on a single 2.5-D system. Our experimental results demonstrate that the HeMu-enabled heterogeneous multichiplet architecture achieves up to $\times 4.8$ in energy efficiency than the homogeneous counterparts.
- 15 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 45 (6) , 2840-2855
HeMu: Energy-Efficient DNN Inferencing via Heterogeneous Multichiplet Architectures
Processing-in-memory (PIM) is a promising solution to accelerate deep neural networks’ (DNNs) inference. However, the area and fabrication costs of monolithic PIM accelerators increase with growing complexity of DNN models. Hence, the design of PIM-based DNN accelerators should leverage the benefits introduced by the emerging chiplet-based 2.5-D architectures. Moreover, chiplets can be designed with various PIM devices, including resistive-random access-memory (ReRAM), ferroelectric field-effect transistor (FeFET), phase change memory (PCM), MRAM, and static random access-memory (SRAM). Each option offers advantages and limitations in terms of power, latency, area, and reliability. None of these PIM-based chiplets has all the desirable features necessary to design high-performance, energy-efficient, and reliable DNN accelerators. Hence, a thorough design space exploration (DSE) is needed to integrate different types of PIM-based chiplets in a single 2.5-D platform. This article proposes a design and optimization framework called HeMu to enable the design of high-performance and energy-efficient multichiplet architectures for accelerating DNN inferencing. HeMu performs a DSE to combine the benefits of more than one type of PIM-based chiplet on a single 2.5-D system. Our experimental results demonstrate that the HeMu-enabled heterogeneous multichiplet architecture achieves up to $\times 4.8$ in energy efficiency than the homogeneous counterparts.
- Open Access
- 26 September 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 24 (5s) , 1-26
THERMOS: Thermally-Aware Multi-Objective Scheduling of AI Workloads on Heterogeneous Multi-Chiplet PIM Architectures
Chiplet-based integration enables large-scale systems that combine diverse technologies, enabling higher yield, lower costs, and scalability, making them well-suited to AI workloads. Processing-in-Memory (PIM) has emerged as a promising solution for AI inference, leveraging technologies such as ReRAM, SRAM, and FeFET, each offering unique advantages and tradeoffs. A heterogeneous chiplet-based PIM architecture can harness the complementary strengths of these technologies to enable higher performance and energy efficiency. However, scheduling AI workloads across such a heterogeneous system is challenging due to competing performance objectives, dynamic workload characteristics, and power and thermal constraints. To address this need, we propose THERMOS, a thermally-aware, multi-objective scheduling framework for AI workloads on heterogeneous multi-chiplet PIM architectures. THERMOS trains a single multi-objective reinforcement learning (MORL) policy that is capable of achieving Pareto-optimal execution time, energy, or a balanced objective at runtime, depending on the target preferences. Comprehensive evaluations show that THERMOS achieves up to 89% faster average execution time and 57% lower average energy consumption than baseline AI workload scheduling algorithms with only 0.14% runtime and 0.022% energy overhead.
- 26 September 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 24 (5s) , 1-26
Open AccessTHERMOS: Thermally-Aware Multi-Objective Scheduling of AI Workloads on Heterogeneous Multi-Chiplet PIM Architectures
Chiplet-based integration enables large-scale systems that combine diverse technologies, enabling higher yield, lower costs, and scalability, making them well-suited to AI workloads. Processing-in-Memory (PIM) has emerged as a promising solution for AI inference, leveraging technologies such as ReRAM, SRAM, and FeFET, each offering unique advantages and tradeoffs. A heterogeneous chiplet-based PIM architecture can harness the complementary strengths of these technologies to enable higher performance and energy efficiency. However, scheduling AI workloads across such a heterogeneous system is challenging due to competing performance objectives, dynamic workload characteristics, and power and thermal constraints. To address this need, we propose THERMOS, a thermally-aware, multi-objective scheduling framework for AI workloads on heterogeneous multi-chiplet PIM architectures. THERMOS trains a single multi-objective reinforcement learning (MORL) policy that is capable of achieving Pareto-optimal execution time, energy, or a balanced objective at runtime, depending on the target preferences. Comprehensive evaluations show that THERMOS achieves up to 89% faster average execution time and 57% lower average energy consumption than baseline AI workload scheduling algorithms with only 0.14% runtime and 0.022% energy overhead.
- Open Access
- 26 September 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 24 (5s) , 1-22
eMamba: Efficient Acceleration Framework for Mamba Models in Edge Computing
State Space Model (SSM)-based machine learning architectures have recently gained significant attention for processing sequential data. Mamba, a recent sequence-to-sequence SSM, offers competitive accuracy with superior computational efficiency compared to state-of-the-art transformer models. While this advantage makes Mamba particularly promising for resource-constrained edge devices, no hardware acceleration frameworks are currently optimized for deploying it in such environments. This article presents eMamba, a comprehensive end-to-end hardware acceleration framework explicitly designed for deploying Mamba models on edge platforms. eMamba maximizes computational efficiency by replacing complex normalization layers with lightweight hardware-aware alternatives and approximating expensive operations, such as SiLU activation and exponentiation, considering the target applications. Then, it performs an approximation-aware neural architecture search (NAS) to tune the learnable parameters used during approximation. Evaluations with Fashion-MNIST, CIFAR-10, and MARS, an open-source human pose estimation dataset, show eMamba achieves comparable accuracy to state-of-the-art techniques using 1.63–19.9× fewer parameters. In addition, it generalizes well to large-scale natural language tasks, demonstrating stable perplexity across varying sequence lengths on the WikiText2 dataset. We also quantize and implement the entire eMamba pipeline on an AMD ZCU102 FPGA and ASIC using GlobalFoundries (GF) 22 nm technology. Experimental results show 4.95–5.62× lower latency and 2.22–9.95× higher throughput, with 4.77× smaller area, 9.84× lower power, and 48.6× lower energy consumption than baseline solutions while maintaining competitive accuracy.
- 26 September 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 24 (5s) , 1-22
Open AccesseMamba: Efficient Acceleration Framework for Mamba Models in Edge Computing
State Space Model (SSM)-based machine learning architectures have recently gained significant attention for processing sequential data. Mamba, a recent sequence-to-sequence SSM, offers competitive accuracy with superior computational efficiency compared to state-of-the-art transformer models. While this advantage makes Mamba particularly promising for resource-constrained edge devices, no hardware acceleration frameworks are currently optimized for deploying it in such environments. This article presents eMamba, a comprehensive end-to-end hardware acceleration framework explicitly designed for deploying Mamba models on edge platforms. eMamba maximizes computational efficiency by replacing complex normalization layers with lightweight hardware-aware alternatives and approximating expensive operations, such as SiLU activation and exponentiation, considering the target applications. Then, it performs an approximation-aware neural architecture search (NAS) to tune the learnable parameters used during approximation. Evaluations with Fashion-MNIST, CIFAR-10, and MARS, an open-source human pose estimation dataset, show eMamba achieves comparable accuracy to state-of-the-art techniques using 1.63–19.9× fewer parameters. In addition, it generalizes well to large-scale natural language tasks, demonstrating stable perplexity across varying sequence lengths on the WikiText2 dataset. We also quantize and implement the entire eMamba pipeline on an AMD ZCU102 FPGA and ASIC using GlobalFoundries (GF) 22 nm technology. Experimental results show 4.95–5.62× lower latency and 2.22–9.95× higher throughput, with 4.77× smaller area, 9.84× lower power, and 48.6× lower energy consumption than baseline solutions while maintaining competitive accuracy.
- 6 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 43 (11) , 4202-4213
Runtime Monitoring of ML-Based Scheduling Algorithms Toward Robust Domain-Specific SoCs
Machine learning (ML) algorithms are being rapidly adopted to perform dynamic resource management tasks in heterogeneous system on chips. For example, ML-based task schedulers can make quick, high-quality decisions at runtime. Like any ML model, these offline-trained policies depend critically on the representative power of the training data. Hence, their performance may diminish or even catastrophically fail under unknown workloads, especially new applications. This article proposes a novel framework to continuously monitor the system to detect unforeseen scenarios using a gradient-based generalization metric called coherence. The proposed framework accurately determines whether the current policy generalizes to new inputs. If not, it incrementally trains the ML scheduler to ensure the robustness of the task-scheduling decisions. The proposed framework is evaluated thoroughly with a domain-specific SoC and six real-world applications. It can detect whether the trained scheduler generalizes to the current workload with 88.75%–98.39% accuracy. Furthermore, it enables $1.1\times -14\times $ faster execution time when the scheduler is incrementally trained. Finally, overhead analysis performed on an Nvidia Jetson Xavier NX board shows that the proposed framework can run as a real-time background task.
- 6 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 43 (11) , 4202-4213
Runtime Monitoring of ML-Based Scheduling Algorithms Toward Robust Domain-Specific SoCs
Machine learning (ML) algorithms are being rapidly adopted to perform dynamic resource management tasks in heterogeneous system on chips. For example, ML-based task schedulers can make quick, high-quality decisions at runtime. Like any ML model, these offline-trained policies depend critically on the representative power of the training data. Hence, their performance may diminish or even catastrophically fail under unknown workloads, especially new applications. This article proposes a novel framework to continuously monitor the system to detect unforeseen scenarios using a gradient-based generalization metric called coherence. The proposed framework accurately determines whether the current policy generalizes to new inputs. If not, it incrementally trains the ML scheduler to ensure the robustness of the task-scheduling decisions. The proposed framework is evaluated thoroughly with a domain-specific SoC and six real-world applications. It can detect whether the trained scheduler generalizes to the current workload with 88.75%–98.39% accuracy. Furthermore, it enables $1.1\times -14\times $ faster execution time when the scheduler is incrementally trained. Finally, overhead analysis performed on an Nvidia Jetson Xavier NX board shows that the proposed framework can run as a real-time background task.