Joyce Cahoon
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
Joyce Cahoon has a total of 67 co-authors. Here are the top co-authors by number of shared publications:
- Yiwen Zhu
Microsoft, San Francisco, USA
4 shared publications - Anna Pavlenko
Microsoft, Redmond, WA, USA
3 shared publications - Andreas Mueller
Microsoft, Mountain View, CA, USA
2 shared publications - Brian Kroth
Microsoft, Madison, WI, USA
2 shared publications - Carlo Curino
Microsoft, Redmond, USA
2 shared publications - Jesús Camacho-Rodríguez
Microsoft, Mountain View, CA, USA
2 shared publications
Latest Publications
- Open Access
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 8809-8817
GraphMind: From Operational Traces to Self-Evolving Workflow Automation
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 8809-8817
Open AccessGraphMind: From Operational Traces to Self-Evolving Workflow Automation
- Open Access
- 22 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-11
Optimizing open-domain question answering with graph-based retrieval augmented generation
- 22 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-11
Open AccessOptimizing open-domain question answering with graph-based retrieval augmented generation
- Open Access
- 9 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 241-254
Vertically Autoscaling Monolithic Applications with CaaSPER: Scalable C ontainer- a s- a - S ervice P erformance E nhanced R esizing Algorithm for the Cloud
Kubernetes has emerged as a prominent open-source platform for managing cloud applications, including stateful databases. These monolithic applications rely on vertical scaling, adjusting CPU cores based on load fluctuations. However, our analysis of Kubernetes-based Database-as-a-Service (DBaaS) offerings at Microsoft revealed that many customers consistently over-provision resources for peak workloads, neglecting cost-saving opportunities through resource scale-down. We found that there is a gap in the ability of existing vertical autoscaling tools to minimize resource slack and respond promptly to throttling, leading to increased costs and impacting crucial metrics such as throughput and availability. To address this challenge, we propose CaaSPER, a vertical autoscaling algorithm that blends reactive and proactive strategies. By dynamically adjusting CPU resources, CaaSPER minimizes resource slack, maintains optimal CPU utilization, and reduces throttling. Importantly, customers have the flexibility to prioritize either cost savings or high performance based on their preferences. Extensive testing demonstrates that CaaSPER effectively reduces throttling and keeps CPU utilization within target levels. CaaSPER is designed to be application-agnostic and platform-agnostic, with potential for extension to other applications requiring vertical autoscaling.
- 9 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 241-254
Open AccessVertically Autoscaling Monolithic Applications with CaaSPER: Scalable C ontainer- a s- a - S ervice P erformance E nhanced R esizing Algorithm for the Cloud
Kubernetes has emerged as a prominent open-source platform for managing cloud applications, including stateful databases. These monolithic applications rely on vertical scaling, adjusting CPU cores based on load fluctuations. However, our analysis of Kubernetes-based Database-as-a-Service (DBaaS) offerings at Microsoft revealed that many customers consistently over-provision resources for peak workloads, neglecting cost-saving opportunities through resource scale-down. We found that there is a gap in the ability of existing vertical autoscaling tools to minimize resource slack and respond promptly to throttling, leading to increased costs and impacting crucial metrics such as throughput and availability. To address this challenge, we propose CaaSPER, a vertical autoscaling algorithm that blends reactive and proactive strategies. By dynamically adjusting CPU resources, CaaSPER minimizes resource slack, maintains optimal CPU utilization, and reduces throttling. Importantly, customers have the flexibility to prioritize either cost savings or high performance based on their preferences. Extensive testing demonstrates that CaaSPER effectively reduces throttling and keeps CPU utilization within target levels. CaaSPER is designed to be application-agnostic and platform-agnostic, with potential for extension to other applications requiring vertical autoscaling.
- Open Access
- 4 June 2023
- Published by Association for Computing Machinery (ACM)
- p. 217-224
Towards Building Autonomous Data Services on Azure
- 4 June 2023
- Published by Association for Computing Machinery (ACM)
- p. 217-224
Open AccessTowards Building Autonomous Data Services on Azure
- 23 December 2020
- Published by Wiley in Quality and Reliability Engineering International
- Vol. 37 (7) , 3108-3124
Practical reliability growth modeling
Reliability growth models are commonly used in the Department of Defense (DoD) to plan, track, and project reliability during system acquisition and testing. We describe two commonly used classes of reliability growth models for continuous failure time data and the metrics appropriate for their use. We also present two Bayesian reliability growth models that are based on the DoD models. The Bayesian models are easily interpretable in a statistical framework, which supports estimation and uncertainty quantification. Our goal is to provide a practical understanding of the development, implementation, and use of reliability growth models across a sequence of DoD testing events.
- 23 December 2020
- Published by Wiley in Quality and Reliability Engineering International
- Vol. 37 (7) , 3108-3124
Practical reliability growth modeling
Reliability growth models are commonly used in the Department of Defense (DoD) to plan, track, and project reliability during system acquisition and testing. We describe two commonly used classes of reliability growth models for continuous failure time data and the metrics appropriate for their use. We also present two Bayesian reliability growth models that are based on the DoD models. The Bayesian models are easily interpretable in a statistical framework, which supports estimation and uncertainty quantification. Our goal is to provide a practical understanding of the development, implementation, and use of reliability growth models across a sequence of DoD testing events.
- 13 August 2019
- Published by Wiley in Veterinary and Comparative Oncology
- Vol. 17 (4) , 562-569
Radiotherapy isocenters verified by matching to bony landmarks of the canine and feline head differ when localized using volumetric versus planar imaging
The “gold standard” for verification of patient positioning before linear accelerator‐based stereotactic radiation therapy is kilovoltage cone‐beam computed tomography (kV‐CBCT), which is not uniformly available or utilized; planar imaging is sometimes used instead. The primary aim of this study was to determine if the position of the bony skull, when used as a surrogate for isocenter verification, is different when orthogonal megavoltage (MV) portal or kilovoltage (kV/kV) radiographs are used for image guidance, rather than kV‐CBCT. A secondary aim was to determine the influence of intra‐observer variability, body size and skull conformation on positioning, as determined using these three imaging modalities. Dogs and cats receiving radiotherapy of the head were recruited for this prospective analytical study. Planar (MV portal and kV/kV images) and volumetric (kV‐CBCT) images were acquired before treatment, and manually coregistered with reference images. Differences in skull position when matched based on MV portal, kV/kV images and kV‐CBCT were compared. A total of 65 subjects and 148 unique datasets were evaluated. The Wilcoxon rank‐sum test was used to evaluate effects of transitioning between imaging modalities. When comparing magnitude of shifts in MV to kV‐CBCT, MV to kV/kV and kV/kV to kV‐CBCT, there were statistically significant differences. Results were not measurably impacted by body size, skull conformation or interobserver differences. Based on shift magnitude and direction, an isotropic setup margin of at least 1 mm should be incorporated within the planning target volume when MV or kV planar imaging is used for position verification.
- 13 August 2019
- Published by Wiley in Veterinary and Comparative Oncology
- Vol. 17 (4) , 562-569
Radiotherapy isocenters verified by matching to bony landmarks of the canine and feline head differ when localized using volumetric versus planar imaging
The “gold standard” for verification of patient positioning before linear accelerator‐based stereotactic radiation therapy is kilovoltage cone‐beam computed tomography (kV‐CBCT), which is not uniformly available or utilized; planar imaging is sometimes used instead. The primary aim of this study was to determine if the position of the bony skull, when used as a surrogate for isocenter verification, is different when orthogonal megavoltage (MV) portal or kilovoltage (kV/kV) radiographs are used for image guidance, rather than kV‐CBCT. A secondary aim was to determine the influence of intra‐observer variability, body size and skull conformation on positioning, as determined using these three imaging modalities. Dogs and cats receiving radiotherapy of the head were recruited for this prospective analytical study. Planar (MV portal and kV/kV images) and volumetric (kV‐CBCT) images were acquired before treatment, and manually coregistered with reference images. Differences in skull position when matched based on MV portal, kV/kV images and kV‐CBCT were compared. A total of 65 subjects and 148 unique datasets were evaluated. The Wilcoxon rank‐sum test was used to evaluate effects of transitioning between imaging modalities. When comparing magnitude of shifts in MV to kV‐CBCT, MV to kV/kV and kV/kV to kV‐CBCT, there were statistically significant differences. Results were not measurably impacted by body size, skull conformation or interobserver differences. Based on shift magnitude and direction, an isotropic setup margin of at least 1 mm should be incorporated within the planning target volume when MV or kV planar imaging is used for position verification.
- 9 August 2019
- Published by The R Foundation
butcher: Model Butcher