Chandra Narayanaswami
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Chandra Narayanaswami has a total of 44 co-authors. Here are the top co-authors by number of shared publications:
- Saurabh Jha
IBM Research, Yorktown Heights, New York, USA
3 shared publications - Archit Patke
University of Illinois Urbana-Champaign, Urbana, USA
2 shared publications - Pavithra Harsha
IBM Research, Yorktown Heights, NY, USA
2 shared publications - Ravishankar K. Iyer
University of Illinois at Urbana-Champaign, Urbana, Illinois, USA
2 shared publications - Umamaheswari Devi
IBM Research, Bangalore, India
2 shared publications - Zbigniew Kalbarczyk
University of Illinois at Urbana-Champaign, Illinois, USA
2 shared publications
Latest Publications
- Open Access
- 1 March 2026
- Published by Association for Computing Machinery (ACM)
Sustainable Computing: State-of-the-Art and Future Outlook
- 1 March 2026
- Published by Association for Computing Machinery (ACM)
Open AccessSustainable Computing: State-of-the-Art and Future Outlook
- Open Access
- 19 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 125-133
CPU-Limits kill Performance: Time to rethink Resource Control
- 19 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 125-133
Open AccessCPU-Limits kill Performance: Time to rethink Resource Control
- Open Access
- 15 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 1145-1164
Story of Two GPUs: Characterizing the Resilience of Hopper H100 and Ampere A100 GPUs
- 15 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 1145-1164
Open AccessStory of Two GPUs: Characterizing the Resilience of Hopper H100 and Ampere A100 GPUs
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 850-859
Practical Contextual Bandits for Large-Scale Structured Discrete Constrained Optimization Problems
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 850-859
Open AccessPractical Contextual Bandits for Large-Scale Structured Discrete Constrained Optimization Problems
- Open Access
- 20 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 18-35
Queue Management for SLO-Oriented Large Language Model Serving
Large language model (LLM) serving is becoming an increasingly critical workload for cloud providers. Existing LLM serving systems focus on interactive requests, such as chatbots and coding assistants, with tight latency SLO requirements. However, when such systems execute batch requests that have relaxed SLOs along with interactive requests, it leads to poor multiplexing and inefficient resource utilization. To address these challenges, we propose QLM, a queue management system for LLM serving. QLM maintains batch and interactive requests across different models and SLOs in a request queue. Optimal ordering of the request queue is critical to maintain SLOs while ensuring high resource utilization. To generate this optimal ordering, QLM uses a Request Waiting Time (RWT) Estimator that estimates the waiting times for requests in the request queue. These estimates are used by a global scheduler to orchestrate LLM Serving Operations (LSOs) such as request pulling, request eviction, load balancing, and model swapping. Evaluation on heterogeneous GPU devices and models with real-world LLM serving dataset shows that QLM improves SLO attainment by 40-90% and throughput by 20-400% while maintaining or improving device utilization compared to other state-of-the-art LLM serving systems. QLM's evaluation is based on the production requirements of a cloud provider. QLM is publicly available at https://www.github.com/QLM-project/QLM.
- 20 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 18-35
Open AccessQueue Management for SLO-Oriented Large Language Model Serving
Large language model (LLM) serving is becoming an increasingly critical workload for cloud providers. Existing LLM serving systems focus on interactive requests, such as chatbots and coding assistants, with tight latency SLO requirements. However, when such systems execute batch requests that have relaxed SLOs along with interactive requests, it leads to poor multiplexing and inefficient resource utilization. To address these challenges, we propose QLM, a queue management system for LLM serving. QLM maintains batch and interactive requests across different models and SLOs in a request queue. Optimal ordering of the request queue is critical to maintain SLOs while ensuring high resource utilization. To generate this optimal ordering, QLM uses a Request Waiting Time (RWT) Estimator that estimates the waiting times for requests in the request queue. These estimates are used by a global scheduler to orchestrate LLM Serving Operations (LSOs) such as request pulling, request eviction, load balancing, and model swapping. Evaluation on heterogeneous GPU devices and models with real-world LLM serving dataset shows that QLM improves SLO attainment by 40-90% and throughput by 20-400% while maintaining or improving device utilization compared to other state-of-the-art LLM serving systems. QLM's evaluation is based on the production requirements of a cloud provider. QLM is publicly available at https://www.github.com/QLM-project/QLM.
- Open Access
- 4 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 891-900
Hierarchical Proxy Modeling for Improved HPO in Time Series Forecasting
Selecting the right set of hyperparameters is crucial in time series forecasting. The classical temporal cross-validation framework for hyperparameter optimization (HPO) often leads to poor test performance because of a possible mismatch between validation and test periods. To address this test-validation mismatch, we propose a novel technique, H-Pro to drive HPO via test proxies by exploiting data hierarchies often associated with time series datasets. Since higher-level aggregated time series often show less irregularity and better predictability as compared to the lowest-level time series which can be sparse and intermittent, we optimize the hyperparameters of the lowest-level base-forecaster by leveraging the proxy forecasts for the test period generated from the forecasters at higher levels. H-Pro can be applied on any off-the-shelf machine learning model to perform HPO. We validate the efficacy of our technique with extensive empirical evaluation on five publicly available hierarchical forecasting datasets. Our approach outperforms existing state-of-the-art methods in Tourism, Wiki, and Traffic datasets, and achieves competitive result in Tourism-L dataset, without any model-specific enhancements. Moreover, our method outperforms the winning method of the M5 forecast accuracy competition.
- 4 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 891-900
Open AccessHierarchical Proxy Modeling for Improved HPO in Time Series Forecasting
Selecting the right set of hyperparameters is crucial in time series forecasting. The classical temporal cross-validation framework for hyperparameter optimization (HPO) often leads to poor test performance because of a possible mismatch between validation and test periods. To address this test-validation mismatch, we propose a novel technique, H-Pro to drive HPO via test proxies by exploiting data hierarchies often associated with time series datasets. Since higher-level aggregated time series often show less irregularity and better predictability as compared to the lowest-level time series which can be sparse and intermittent, we optimize the hyperparameters of the lowest-level base-forecaster by leveraging the proxy forecasts for the test period generated from the forecasters at higher levels. H-Pro can be applied on any off-the-shelf machine learning model to perform HPO. We validate the efficacy of our technique with extensive empirical evaluation on five publicly available hierarchical forecasting datasets. Our approach outperforms existing state-of-the-art methods in Tourism, Wiki, and Traffic datasets, and achieves competitive result in Tourism-L dataset, without any model-specific enhancements. Moreover, our method outperforms the winning method of the M5 forecast accuracy competition.
- Open Access
- 9 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Towards Carbon Footprint Management in Hybrid Multicloud
- 9 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessTowards Carbon Footprint Management in Hybrid Multicloud
- 1 October 2002
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Pervasive Computing
- Vol. 1 (4) , 42-48
Designing a new form factor for wearable computing
Despite significant improvements in underlying technologies, the wearable computing field is still in its youth. The article offers several methods that can help accelerate the process from vision to product.
- 1 October 2002
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Pervasive Computing
- Vol. 1 (4) , 42-48
Designing a new form factor for wearable computing
Despite significant improvements in underlying technologies, the wearable computing field is still in its youth. The article offers several methods that can help accelerate the process from vision to product.