Rutwik Jain
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
Rutwik Jain has a total of 45 co-authors. Here are the top co-authors by number of shared publications:
- Matthew D. Sinclair
University of Wisconsin–Madison, Madison, WI, USA
7 shared publications - Shivaram Venkataraman
University of Wisconsin-Madison, Madison, WI, USA
6 shared publications - Yiwei Jiang
University of Wisconsin-Madison, Madison, WI, USA
3 shared publications - Geoffrey Fox
University of Virginia
2 shared publications - Gregor von Laszewski
Biocomplexity Institute and Initiative University of Virginia 22911 Charlottesville VA USA
2 shared publications Marco Colombo
2 shared publications
Latest Publications
- Open Access
- 20 July 2026
- Published by Association for Computing Machinery (ACM) in ACM SIGMETRICS Performance Evaluation Review
- Vol. 54 (1) , 117-119
Minos: Systematically Classifying Performance and Power Characteristics of GPU Workloads on HPC Clusters
As large-scale HPC compute clusters increasingly adopt accelerators such as GPUs to meet the voracious demands of modern workloads, these clusters are increasingly becoming power constrained. Given that modern applications can often temporarily exceed the power ratings of the accelerators (''power spikes''), current and future HPC systems must optimize for both power and performance together. However, this is made difficult by increasingly diverse applications, which often require bespoke optimizations to run efficiently on each cluster. Traditionally researchers overcome this problem by profiling applications on specific clusters and optimizing, but the scale, algorithmic diversity, and lack of effective tools make this challenging. To overcome these inefficiencies, we propose Minos, a systematic classification mechanism that identifies similar application characteristics via low-cost profiling for power and performance. This allows us to group similarly behaving workloads into a finite number of distinct classes and reduce the overhead of extensively profiling new workloads. For example, when predicting frequency capping behavior for a previously unseen application, Minos reduces profiling time by 89%. Moreover, across 18 popular graph analytics, HPC, HPC+ML, and ML workloads, Minos achieves a mean error of 4% for power predictions and 3% for performance predictions, significantly improving predictions over state-of-the-art approaches by 10%.
- 20 July 2026
- Published by Association for Computing Machinery (ACM) in ACM SIGMETRICS Performance Evaluation Review
- Vol. 54 (1) , 117-119
Open AccessMinos: Systematically Classifying Performance and Power Characteristics of GPU Workloads on HPC Clusters
As large-scale HPC compute clusters increasingly adopt accelerators such as GPUs to meet the voracious demands of modern workloads, these clusters are increasingly becoming power constrained. Given that modern applications can often temporarily exceed the power ratings of the accelerators (''power spikes''), current and future HPC systems must optimize for both power and performance together. However, this is made difficult by increasingly diverse applications, which often require bespoke optimizations to run efficiently on each cluster. Traditionally researchers overcome this problem by profiling applications on specific clusters and optimizing, but the scale, algorithmic diversity, and lack of effective tools make this challenging. To overcome these inefficiencies, we propose Minos, a systematic classification mechanism that identifies similar application characteristics via low-cost profiling for power and performance. This allows us to group similarly behaving workloads into a finite number of distinct classes and reduce the overhead of extensively profiling new workloads. For example, when predicting frequency capping behavior for a previously unseen application, Minos reduces profiling time by 89%. Moreover, across 18 popular graph analytics, HPC, HPC+ML, and ML workloads, Minos achieves a mean error of 4% for power predictions and 3% for performance predictions, significantly improving predictions over state-of-the-art approaches by 10%.
- Open Access
- 8 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 117-119
Minos: Systematically Classifying Performance and Power Characteristics of GPU Workloads on HPC Clusters
As large-scale HPC compute clusters increasingly adopt accelerators such as GPUs to meet the voracious demands of modern workloads, these clusters are increasingly becoming power constrained. Given that modern applications can often temporarily exceed the power ratings of the accelerators (''power spikes''), current and future HPC systems must optimize for both power and performance together. However, this is made difficult by increasingly diverse applications, which often require bespoke optimizations to run efficiently on each cluster. Traditionally researchers overcome this problem by profiling applications on specific clusters and optimizing, but the scale, algorithmic diversity, and lack of effective tools make this challenging. To overcome these inefficiencies, we propose Minos, a systematic classification mechanism that identifies similar application characteristics via low-cost profiling for power and performance. This allows us to group similarly behaving workloads into a finite number of distinct classes and reduce the overhead of extensively profiling new workloads. For example, when predicting frequency capping behavior for a previously unseen application, Minos reduces profiling time by 89%. Moreover, across 18 popular graph analytics, HPC, HPC+ML, and ML workloads, Minos achieves a mean error of 4% for power predictions and 3% for performance predictions, significantly improving predictions over state-of-the-art approaches by 10%.
- 8 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 117-119
Open AccessMinos: Systematically Classifying Performance and Power Characteristics of GPU Workloads on HPC Clusters
As large-scale HPC compute clusters increasingly adopt accelerators such as GPUs to meet the voracious demands of modern workloads, these clusters are increasingly becoming power constrained. Given that modern applications can often temporarily exceed the power ratings of the accelerators (''power spikes''), current and future HPC systems must optimize for both power and performance together. However, this is made difficult by increasingly diverse applications, which often require bespoke optimizations to run efficiently on each cluster. Traditionally researchers overcome this problem by profiling applications on specific clusters and optimizing, but the scale, algorithmic diversity, and lack of effective tools make this challenging. To overcome these inefficiencies, we propose Minos, a systematic classification mechanism that identifies similar application characteristics via low-cost profiling for power and performance. This allows us to group similarly behaving workloads into a finite number of distinct classes and reduce the overhead of extensively profiling new workloads. For example, when predicting frequency capping behavior for a previously unseen application, Minos reduces profiling time by 89%. Moreover, across 18 popular graph analytics, HPC, HPC+ML, and ML workloads, Minos achieves a mean error of 4% for power predictions and 3% for performance predictions, significantly improving predictions over state-of-the-art approaches by 10%.
- Open Access
- 29 May 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Measurement and Analysis of Computing Systems
- Vol. 10 (2) , 1-27
Minos: Systematically Classifying Performance and Power Characteristics of GPU Workloads on HPC Clusters
As large-scale HPC compute clusters increasingly adopt accelerators such as GPUs to meet the voracious demands of modern workloads, these clusters are increasingly becoming power constrained. Given that modern applications can often temporarily exceed the power ratings of the accelerators (''power spikes''), current and future HPC systems must optimize for both power and performance together. However, this is made difficult by increasingly diverse applications, which often require bespoke optimizations to run efficiently on each cluster. Traditionally researchers overcome this problem by profiling applications on specific clusters and optimizing, but the scale, algorithmic diversity, and lack of effective tools make this challenging. To overcome these inefficiencies, we propose Minos, a systematic classification mechanism that identifies similar application characteristics via low-cost profiling for power and performance. This allows us to group similarly behaving workloads into a finite number of distinct classes and reduce the overhead of extensively profiling new workloads. For example, when predicting frequency capping behavior for a previously unseen application, Minos reduces profiling time by 89%. Moreover, across 18 popular graph analytics, HPC, HPC+ML, and ML workloads, Minos achieves a mean error of 4% for power predictions and 3% for performance predictions, significantly improving predictions over state-of-the-art approaches by 10%.
- 29 May 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Measurement and Analysis of Computing Systems
- Vol. 10 (2) , 1-27
Open AccessMinos: Systematically Classifying Performance and Power Characteristics of GPU Workloads on HPC Clusters
As large-scale HPC compute clusters increasingly adopt accelerators such as GPUs to meet the voracious demands of modern workloads, these clusters are increasingly becoming power constrained. Given that modern applications can often temporarily exceed the power ratings of the accelerators (''power spikes''), current and future HPC systems must optimize for both power and performance together. However, this is made difficult by increasingly diverse applications, which often require bespoke optimizations to run efficiently on each cluster. Traditionally researchers overcome this problem by profiling applications on specific clusters and optimizing, but the scale, algorithmic diversity, and lack of effective tools make this challenging. To overcome these inefficiencies, we propose Minos, a systematic classification mechanism that identifies similar application characteristics via low-cost profiling for power and performance. This allows us to group similarly behaving workloads into a finite number of distinct classes and reduce the overhead of extensively profiling new workloads. For example, when predicting frequency capping behavior for a previously unseen application, Minos reduces profiling time by 89%. Moreover, across 18 popular graph analytics, HPC, HPC+ML, and ML workloads, Minos achieves a mean error of 4% for power predictions and 3% for performance predictions, significantly improving predictions over state-of-the-art approaches by 10%.
- Open Access
AI Benchmark Democratization and Carpentry
Benchmarks are a cornerstone of modern machine learning, enabling reproducibility, comparison, and scientific progress. However, AI benchmarks are increasingly complex, requiring dynamic, AI-focused workflows. Rapid evolution in model architectures, scale, datasets, and deployment contexts makes evaluation a moving target. Large language models often memorize static benchmarks, causing a gap between benchmark results and real-world performance. Beyond traditional static benchmarks, continuous adaptive benchmarking frameworks are needed to align scientific assessment with deployment risks. This calls for skills and education in AI Benchmark Carpentry. From our experience with MLCommons, educational initiatives, and programs like the DOE's Trillion Parameter Consortium, key barriers include high resource demands, limited access to specialized hardware, lack of benchmark design expertise, and uncertainty in relating results to application domains. Current benchmarks often emphasize peak performance on top-tier hardware, offering limited guidance for diverse, real-world scenarios. Benchmarking must become dynamic, incorporating evolving models, updated data, and heterogeneous platforms while maintaining transparency, reproducibility, and interpretability. Democratization requires both technical innovation and systematic education across levels, building sustained expertise in benchmark design and use. Benchmarks should support application-relevant comparisons, enabling informed, context-sensitive decisions. Dynamic, inclusive benchmarking will ensure evaluation keeps pace with AI evolution and supports responsible, reproducible, and accessible AI deployment. Community efforts can provide a foundation for AI Benchmark Carpentry.
Open AccessAI Benchmark Democratization and Carpentry
Benchmarks are a cornerstone of modern machine learning, enabling reproducibility, comparison, and scientific progress. However, AI benchmarks are increasingly complex, requiring dynamic, AI-focused workflows. Rapid evolution in model architectures, scale, datasets, and deployment contexts makes evaluation a moving target. Large language models often memorize static benchmarks, causing a gap between benchmark results and real-world performance. Beyond traditional static benchmarks, continuous adaptive benchmarking frameworks are needed to align scientific assessment with deployment risks. This calls for skills and education in AI Benchmark Carpentry. From our experience with MLCommons, educational initiatives, and programs like the DOE's Trillion Parameter Consortium, key barriers include high resource demands, limited access to specialized hardware, lack of benchmark design expertise, and uncertainty in relating results to application domains. Current benchmarks often emphasize peak performance on top-tier hardware, offering limited guidance for diverse, real-world scenarios. Benchmarking must become dynamic, incorporating evolving models, updated data, and heterogeneous platforms while maintaining transparency, reproducibility, and interpretability. Democratization requires both technical innovation and systematic education across levels, building sustained expertise in benchmark design and use. Benchmarks should support application-relevant comparisons, enabling informed, context-sensitive decisions. Dynamic, inclusive benchmarking will ensure evaluation keeps pace with AI evolution and supports responsible, reproducible, and accessible AI deployment. Community efforts can provide a foundation for AI Benchmark Carpentry.
- Open Access
An MLCommons Scientific Benchmarks Ontology
Scientific machine learning research spans diverse domains and data modalities, yet existing benchmark efforts remain siloed and lack standardization. This makes novel and transformative applications of machine learning to critical scientific use-cases more fragmented and less clear in pathways to impact. This paper introduces an ontology for scientific benchmarking developed through a unified, community-driven effort that extends the MLCommons ecosystem to cover physics, chemistry, materials science, biology, climate science, and more. Building on prior initiatives such as XAI-BENCH, FastML Science Benchmarks, PDEBench, and the SciMLBench framework, our effort consolidates a large set of disparate benchmarks and frameworks into a single taxonomy of scientific, application, and system-level benchmarks. New benchmarks can be added through an open submission workflow coordinated by the MLCommons Science Working Group and evaluated against a six-category rating rubric that promotes and identifies high-quality benchmarks, enabling stakeholders to select benchmarks that meet their specific needs. The architecture is extensible, supporting future scientific and AI/ML motifs, and we discuss methods for identifying emerging computing patterns for unique scientific workloads. The MLCommons Science Benchmarks Ontology provides a standardized, scalable foundation for reproducible, cross-domain benchmarking in scientific machine learning. A companion webpage for this work has also been developed as the effort evolves: https://mlcommons-science.github.io/benchmark/
Open AccessAn MLCommons Scientific Benchmarks Ontology
Scientific machine learning research spans diverse domains and data modalities, yet existing benchmark efforts remain siloed and lack standardization. This makes novel and transformative applications of machine learning to critical scientific use-cases more fragmented and less clear in pathways to impact. This paper introduces an ontology for scientific benchmarking developed through a unified, community-driven effort that extends the MLCommons ecosystem to cover physics, chemistry, materials science, biology, climate science, and more. Building on prior initiatives such as XAI-BENCH, FastML Science Benchmarks, PDEBench, and the SciMLBench framework, our effort consolidates a large set of disparate benchmarks and frameworks into a single taxonomy of scientific, application, and system-level benchmarks. New benchmarks can be added through an open submission workflow coordinated by the MLCommons Science Working Group and evaluated against a six-category rating rubric that promotes and identifies high-quality benchmarks, enabling stakeholders to select benchmarks that meet their specific needs. The architecture is extensible, supporting future scientific and AI/ML motifs, and we discuss methods for identifying emerging computing patterns for unique scientific workloads. The MLCommons Science Benchmarks Ontology provides a standardized, scalable foundation for reproducible, cross-domain benchmarking in scientific machine learning. A companion webpage for this work has also been developed as the effort evolves: https://mlcommons-science.github.io/benchmark/
- 10 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 43 (6) , 109-124
Fifty Years of the International Symposium on Computer Architecture: A Data-Driven Retrospective
2023 marked the fiftieth year of the International Symposium on Computer Architecture (ISCA). As one of the oldest and preeminent computer architecture conferences, ISCA represents a microcosm of the broader community; correspondingly, a 50-year-retrospective offers us a great way to track the impact and evolution of the field. Analyzing the content and impact of all the papers published at ISCA so far, we show how computer architecture research has been at the forefront of advances that have driven the broader computing ecosystem. Decadal trends show a dynamic and rapidly-evolving field, with diverse contributions. Examining how the most highly-cited papers achieve their popularity reveals interesting trends on technology adoption curves and the path to impact. Our data also highlights a growing and thriving community, with interesting insights on diversity and scale. We conclude with a summary of the celebratory panel held at ISCA, with observations on the exciting future ahead.
- 10 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 43 (6) , 109-124
Fifty Years of the International Symposium on Computer Architecture: A Data-Driven Retrospective
2023 marked the fiftieth year of the International Symposium on Computer Architecture (ISCA). As one of the oldest and preeminent computer architecture conferences, ISCA represents a microcosm of the broader community; correspondingly, a 50-year-retrospective offers us a great way to track the impact and evolution of the field. Analyzing the content and impact of all the papers published at ISCA so far, we show how computer architecture research has been at the forefront of advances that have driven the broader computing ecosystem. Decadal trends show a dynamic and rapidly-evolving field, with diverse contributions. Examining how the most highly-cited papers achieve their popularity reveals interesting trends on technology adoption curves and the path to impact. Our data also highlights a growing and thriving community, with interesting insights on diversity and scale. We conclude with a summary of the celebratory panel held at ISCA, with observations on the exciting future ahead.
- Open Access
Not All GPUs Are Created Equal: Characterizing Variability in Large-Scale, Accelerator-Rich Systems
Scientists are increasingly exploring and utilizing the massive parallelism of general-purpose accelerators such as GPUs for scientific breakthroughs. As a result, datacenters, hyperscalers, national computing centers, and supercomputers have procured hardware to support this evolving application paradigm. These systems contain hundreds to tens of thousands of accelerators, enabling peta- and exa-scale levels of compute for scientific workloads. Recent work demonstrated that power management (PM) can impact application performance in CPU-based HPC systems, even when machines have the same architecture and SKU (stock keeping unit). This variation occurs due to manufacturing variability and the chip's PM. However, while modern HPC systems widely employ accelerators such as GPUs, it is unclear how much this variability affects applications. Accordingly, we seek to characterize the extent of variation due to GPU PM in modern HPC and supercomputing systems. We study a variety of applications that stress different GPU components on five large-scale computing centers with modern GPUs: Oak Ridge's Summit, Sandia's Vortex, TACC's Frontera and Longhorn, and Livermore's Corona. These clusters use a variety of cooling methods and GPU vendors. In total, we collect over 18,800 hours of data across more than 90% of the GPUs in these clusters. Regardless of the application, cluster, GPU vendor, and cooling method, our results show significant variation: 8% (max 22%) average performance variation even though the GPU architecture and vendor SKU are identical within each cluster, with outliers up to 1.5X slower than the median GPU. These results highlight the difficulty in efficiently using existing GPU clusters for modern HPC and scientific workloads, and the need to embrace variability in future accelerator-based systems.
Open AccessNot All GPUs Are Created Equal: Characterizing Variability in Large-Scale, Accelerator-Rich Systems
Scientists are increasingly exploring and utilizing the massive parallelism of general-purpose accelerators such as GPUs for scientific breakthroughs. As a result, datacenters, hyperscalers, national computing centers, and supercomputers have procured hardware to support this evolving application paradigm. These systems contain hundreds to tens of thousands of accelerators, enabling peta- and exa-scale levels of compute for scientific workloads. Recent work demonstrated that power management (PM) can impact application performance in CPU-based HPC systems, even when machines have the same architecture and SKU (stock keeping unit). This variation occurs due to manufacturing variability and the chip's PM. However, while modern HPC systems widely employ accelerators such as GPUs, it is unclear how much this variability affects applications. Accordingly, we seek to characterize the extent of variation due to GPU PM in modern HPC and supercomputing systems. We study a variety of applications that stress different GPU components on five large-scale computing centers with modern GPUs: Oak Ridge's Summit, Sandia's Vortex, TACC's Frontera and Longhorn, and Livermore's Corona. These clusters use a variety of cooling methods and GPU vendors. In total, we collect over 18,800 hours of data across more than 90% of the GPUs in these clusters. Regardless of the application, cluster, GPU vendor, and cooling method, our results show significant variation: 8% (max 22%) average performance variation even though the GPU architecture and vendor SKU are identical within each cluster, with outliers up to 1.5X slower than the median GPU. These results highlight the difficulty in efficiently using existing GPU clusters for modern HPC and scientific workloads, and the need to embrace variability in future accelerator-based systems.
- 1 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 871-879
Modelling and Simulation of Onboard Wire Antennas for a 3U CubeSat
This paper discusses the results obtained and conclusions made from modelling and simulation of the on-board antennas for a 3U CubeSat being developed by Team Anant, a group of undergraduate students at BITS Pilani. The nanosatellite features a hyperspectral camera as its primary payload. The paper initially documents the reasons for selection of a turnstile antenna for downlink in the 435 MHz band of the UHF region, and a monopole for uplink in the 144 MHz band of the VHF region. Ansys High Frequency Structural Simulator (HFSS) has been used to carry out parametric simulations for desirable network matching, radiation pattern and directivity. Further, the effect of the metallic structure of the CubeSat is examined for optimization of antenna parameters. The rationale for arriving at two particular positions for the antennas on the CubeSat is elucidated based on electromagnetic compatibility (EMC). Two different positions for the antennas are arrived at and a comparative analysis is provided. Simulations and results are also presented on the mutual interaction of the two different antennas. It is important to design, model and simulate an antenna using electromagnetic CAD tools, especially for low cost CubeSat projects. The paper is an attempt to provide nano-satellite missions with a systematic strategy for modelling and analysing wire antennas.
- 1 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 871-879
Modelling and Simulation of Onboard Wire Antennas for a 3U CubeSat
This paper discusses the results obtained and conclusions made from modelling and simulation of the on-board antennas for a 3U CubeSat being developed by Team Anant, a group of undergraduate students at BITS Pilani. The nanosatellite features a hyperspectral camera as its primary payload. The paper initially documents the reasons for selection of a turnstile antenna for downlink in the 435 MHz band of the UHF region, and a monopole for uplink in the 144 MHz band of the VHF region. Ansys High Frequency Structural Simulator (HFSS) has been used to carry out parametric simulations for desirable network matching, radiation pattern and directivity. Further, the effect of the metallic structure of the CubeSat is examined for optimization of antenna parameters. The rationale for arriving at two particular positions for the antennas on the CubeSat is elucidated based on electromagnetic compatibility (EMC). Two different positions for the antennas are arrived at and a comparative analysis is provided. Simulations and results are also presented on the mutual interaction of the two different antennas. It is important to design, model and simulate an antenna using electromagnetic CAD tools, especially for low cost CubeSat projects. The paper is an attempt to provide nano-satellite missions with a systematic strategy for modelling and analysing wire antennas.
- 4 October 2019
- Published by Wiley in Microwave and Optical Technology Letters
- Vol. 62 (1) , 484-490
Magnetic quadrupole mode excitation of a cylindrical dielectric resonator antenna using planar feed
In this article, excitation of the magnetic quadrupole mode of a cylindrical dielectric resonator antenna, designated as the HEM21δ mode, using planar feeding technique is discussed for the first time. Two types of planar feeds—a dual‐microstrip feed and a dual‐slot feed are numerically investigated. Results show that the dual‐slot feed is more suitable for exciting the HEM21δ mode as it gives better symmetry of the radiation pattern and lower cross‐polarization level. A DRA prototype (dielectric constant, εr = 24, loss tangent, tanδ = 0.002, diameter, 2a = 19.43 mm and height, h = 7.3 mm) using dual‐slot feed is fabricated and characterized. Measurements demonstrate a resonant frequency of ~ 4.96 GHz, bandwidth of ~4.4%, peak gain of ~3.45 dBi and peak cross‐polar level of ~ − 11 dB for the HEM21δ mode, which are in decent agreement with numerical predictions.
- 4 October 2019
- Published by Wiley in Microwave and Optical Technology Letters
- Vol. 62 (1) , 484-490
Magnetic quadrupole mode excitation of a cylindrical dielectric resonator antenna using planar feed
In this article, excitation of the magnetic quadrupole mode of a cylindrical dielectric resonator antenna, designated as the HEM21δ mode, using planar feeding technique is discussed for the first time. Two types of planar feeds—a dual‐microstrip feed and a dual‐slot feed are numerically investigated. Results show that the dual‐slot feed is more suitable for exciting the HEM21δ mode as it gives better symmetry of the radiation pattern and lower cross‐polarization level. A DRA prototype (dielectric constant, εr = 24, loss tangent, tanδ = 0.002, diameter, 2a = 19.43 mm and height, h = 7.3 mm) using dual‐slot feed is fabricated and characterized. Measurements demonstrate a resonant frequency of ~ 4.96 GHz, bandwidth of ~4.4%, peak gain of ~3.45 dBi and peak cross‐polar level of ~ − 11 dB for the HEM21δ mode, which are in decent agreement with numerical predictions.