Christopher Gill
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Christopher Gill has a total of 63 co-authors. Here are the top co-authors by number of shared publications:
- Chenyang Lu
Department of Computer Science and Engineering, Washington University in St. Louis, St. Louis, Missouri, USA
6 shared publications - An Zou
Shanghai Jiao Tong University
3 shared publications - Douglas C. Schmidt
William & Mary, Computer Science, Williamsburg, Virginia, United States
3 shared publications - Xuan Zhang
Northeastern University, Boston, MA, USA
3 shared publications - Chong Li
Department of Computer Science and Engineering, Washington University in St. Louis, St. Louis, MO, USA
2 shared publications - Insup Lee
University of Pennsylvania
2 shared publications
Latest Publications
- Open Access
Measure-Theoretic Diagnostics of Architectural Entanglement in Asymmetric Multiprocessing Systems: A Boltzmann Uniqueness Approach
Orchestration of Asymmetric Multiprocessing Platforms (AMPs), such as ARM big.LITTLE, frequently relies on the heuristic assumption of cluster independence, wherein high-performance (“Big”) and high-efficiency (“Little”) cores operate as computationally orthogonal resources. These cores are partitioned into “islands” of separate power/performance clusters, operating on independent/voltage frequency rails. However, these platforms share resources, including Last-Level Cache (LLC), main memory, and interconnects across all cores. Therefore, we assume that islands interact, operating in a functionally “coupled state.” To conduct a measure-theoretic evaluation of this assumption, we apply the Boltzmann uniqueness theorem, recently demonstrated to be the singular method to determine the veracity of this assumption. Mathematically, we define an “uncoupled” system as one whose joint resource measurement is strictly the convolution of its subsystem measures. We evaluate two distinct AMP topologies—Orange Pi 5 and Cubie A7A under controlled saturation—and demonstrate a systemic failure of convolution commutativity. We subsequently expand this investigation to high-performance x86 hybrid architectures via the Intel i7-12800H platform. Our findings, characterized by significant negative power correlations and the failure of predictive convolution models, constitute a counterexample for cluster independence. We identify shared architectural resources, specifically the LLC and shared power rails, as the likely physical mechanisms of “architectural entanglement,” rendering traditional additive performance models underspecified.
Open AccessMeasure-Theoretic Diagnostics of Architectural Entanglement in Asymmetric Multiprocessing Systems: A Boltzmann Uniqueness Approach
Orchestration of Asymmetric Multiprocessing Platforms (AMPs), such as ARM big.LITTLE, frequently relies on the heuristic assumption of cluster independence, wherein high-performance (“Big”) and high-efficiency (“Little”) cores operate as computationally orthogonal resources. These cores are partitioned into “islands” of separate power/performance clusters, operating on independent/voltage frequency rails. However, these platforms share resources, including Last-Level Cache (LLC), main memory, and interconnects across all cores. Therefore, we assume that islands interact, operating in a functionally “coupled state.” To conduct a measure-theoretic evaluation of this assumption, we apply the Boltzmann uniqueness theorem, recently demonstrated to be the singular method to determine the veracity of this assumption. Mathematically, we define an “uncoupled” system as one whose joint resource measurement is strictly the convolution of its subsystem measures. We evaluate two distinct AMP topologies—Orange Pi 5 and Cubie A7A under controlled saturation—and demonstrate a systemic failure of convolution commutativity. We subsequently expand this investigation to high-performance x86 hybrid architectures via the Intel i7-12800H platform. Our findings, characterized by significant negative power correlations and the failure of predictive convolution models, constitute a counterexample for cluster independence. We identify shared architectural resources, specifically the LLC and shared power rails, as the likely physical mechanisms of “architectural entanglement,” rendering traditional additive performance models underspecified.
- Open Access
- 6 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 218-228
Elastic Scheduling for Graceful Degradation of Mixed-Criticality Systems
- 6 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 218-228
Open AccessElastic Scheduling for Graceful Degradation of Mixed-Criticality Systems
- 9 February 2023
- Published by Springer Nature
- p. 212-230
Reward Delay Attacks on Deep Reinforcement Learning
- 9 February 2023
- Published by Springer Nature
- p. 212-230
Reward Delay Attacks on Deep Reinforcement Learning
- 9 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 34 (5) , 1450-1465
RTGPU: Real-Time GPU Scheduling of Hard Deadline Parallel Tasks With Fine-Grain Utilization
Many emerging cyber-physical systems, such as autonomous vehicles and robots, rely heavily on artificial intelligence and machine learning algorithms to perform important system operations. Since these highly parallel applications are computationally intensive, they need to be accelerated by graphics processing units (GPUs) to meet stringent timing constraints. However, despite the wide adoption of GPUs, efficiently scheduling multiple GPU applications while providing rigorous real-time guarantees remains challenging. Each GPU application has multiple CPU execution and memory copy segments, with GPU kernels running on different hardware resources. Because of the complicated interactions between heterogeneous segments of parallel tasks, high schedulability is hard to achieve with conventional approaches. This paper proposes RTGPU, which combines fine-grain GPU partitioning on the system-side with a novel scheduling algorithm on the theory-side. We start by building a model for CPU and memory copy segments. Leveraging persistent threads, we then implement fine-grained GPU partitioning with improved performance through interleaved execution. To reap the benefits of fine-grained GPU partitioning and schedule multiple parallel GPU applications, we propose a novel real-time scheduling algorithm based on federated scheduling and grid search with uniprocessor fixed-priority scheduling. Our approach provides real-time guarantees to meet hard deadlines and achieves over 11% improvement in system throughput and up to 57% schedulability improvement compared with previous work. We validate and evaluate RTGPU on NVIDIA GPU systems. Our system-side techniques can be applied on mainstream GPUs, and the proposed scheduling theory can be used in general heterogeneous computing platforms which have a similar task execution pattern.
- 9 January 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 34 (5) , 1450-1465
RTGPU: Real-Time GPU Scheduling of Hard Deadline Parallel Tasks With Fine-Grain Utilization
Many emerging cyber-physical systems, such as autonomous vehicles and robots, rely heavily on artificial intelligence and machine learning algorithms to perform important system operations. Since these highly parallel applications are computationally intensive, they need to be accelerated by graphics processing units (GPUs) to meet stringent timing constraints. However, despite the wide adoption of GPUs, efficiently scheduling multiple GPU applications while providing rigorous real-time guarantees remains challenging. Each GPU application has multiple CPU execution and memory copy segments, with GPU kernels running on different hardware resources. Because of the complicated interactions between heterogeneous segments of parallel tasks, high schedulability is hard to achieve with conventional approaches. This paper proposes RTGPU, which combines fine-grain GPU partitioning on the system-side with a novel scheduling algorithm on the theory-side. We start by building a model for CPU and memory copy segments. Leveraging persistent threads, we then implement fine-grained GPU partitioning with improved performance through interleaved execution. To reap the benefits of fine-grained GPU partitioning and schedule multiple parallel GPU applications, we propose a novel real-time scheduling algorithm based on federated scheduling and grid search with uniprocessor fixed-priority scheduling. Our approach provides real-time guarantees to meet hard deadlines and achieves over 11% improvement in system throughput and up to 57% schedulability improvement compared with previous work. We validate and evaluate RTGPU on NVIDIA GPU systems. Our system-side techniques can be applied on mainstream GPUs, and the proposed scheduling theory can be used in general heterogeneous computing platforms which have a similar task execution pattern.
- 18 May 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 42 (2) , 616-629
F-LEMMA: Fast Learning-Based Energy Management for Multi-/Many-Core Processors
Over the last two decades, as microprocessors have evolved to achieve higher computational performance, their power density has also increased at an accelerated rate. Improving energy efficiency and reducing power consumption are therefore critically important to modern computing systems. One effective technique for improving energy efficiency is dynamic voltage and frequency scaling (DVFS). With the emergence of integrated voltage regulators (IVRs), the speed of DVFS can reach microsecond ( $\mu \text{s}$ ) timescales. However, a practical and effective strategy to guide fast DVFS remains a challenge. In this article, we propose F-LEMMA: a fast, learning-based, hierarchical DVFS framework consisting of a global power allocator in the kernel space, a reinforcement learning-based power management scheme at the architecture level, and a swift controller at the digital circuit level. This hierarchical approach leverages computation at the system and architecture levels with the short response time of the swift controller to achieve effective and rapid $\mu \text{s}$ -level power management supported by the IVR. Our experimental results demonstrate that F-LEMMA can achieve significant energy savings (35.2%) across a broad range of workloads. Conservatively compared with existing state-of-the-art DVFS-based power management schemes that can only operate at millisecond timescales, F-LEMMA can provide notable (up to 11%) energy-delay product (EDP) improvements across benchmarks. Compared with state-of-the-art nonlearning-based power management, our method has a universally positive effect on evaluated benchmarks, proving its adaptability.
- 18 May 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 42 (2) , 616-629
F-LEMMA: Fast Learning-Based Energy Management for Multi-/Many-Core Processors
Over the last two decades, as microprocessors have evolved to achieve higher computational performance, their power density has also increased at an accelerated rate. Improving energy efficiency and reducing power consumption are therefore critically important to modern computing systems. One effective technique for improving energy efficiency is dynamic voltage and frequency scaling (DVFS). With the emergence of integrated voltage regulators (IVRs), the speed of DVFS can reach microsecond ( $\mu \text{s}$ ) timescales. However, a practical and effective strategy to guide fast DVFS remains a challenge. In this article, we propose F-LEMMA: a fast, learning-based, hierarchical DVFS framework consisting of a global power allocator in the kernel space, a reinforcement learning-based power management scheme at the architecture level, and a swift controller at the digital circuit level. This hierarchical approach leverages computation at the system and architecture levels with the short response time of the swift controller to achieve effective and rapid $\mu \text{s}$ -level power management supported by the IVR. Our experimental results demonstrate that F-LEMMA can achieve significant energy savings (35.2%) across a broad range of workloads. Conservatively compared with existing state-of-the-art DVFS-based power management schemes that can only operate at millisecond timescales, F-LEMMA can provide notable (up to 11%) energy-delay product (EDP) improvements across benchmarks. Compared with state-of-the-art nonlearning-based power management, our method has a universally positive effect on evaluated benchmarks, proving its adaptability.
- 29 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Networking
- Vol. 30 (1) , 257-270
Virtualization-Aware Traffic Control for Soft Real-Time Network Traffic on Xen
As the role of virtualization technology becomes more prevalent, the range of applications deployed in virtualized systems is steadily growing. This increasingly includes applications with soft real-time requirements that benefit from low and predictable latency, even when co-located with other virtualized hosts with arbitrary traffic patterns. In this paper, we examine the policies and mechanisms affecting communication latency between virtual machines based on the Xen platform, and identify limitations that can result in long or unpredictable network stack latency for virtual machines deployed on this platform. To address these limitations, we propose and implement VATC , a Virtualization-Aware Traffic Control framework that supports differentiation (via rate-limited prioritization) of outbound and inbound network traffic from co-located virtualized hosts. Results of our experiments show how and why VATC can offer predictable (soft) latency guarantees to applications running on virtualized hosts with minimum overhead.
- 29 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Networking
- Vol. 30 (1) , 257-270
Virtualization-Aware Traffic Control for Soft Real-Time Network Traffic on Xen
As the role of virtualization technology becomes more prevalent, the range of applications deployed in virtualized systems is steadily growing. This increasingly includes applications with soft real-time requirements that benefit from low and predictable latency, even when co-located with other virtualized hosts with arbitrary traffic patterns. In this paper, we examine the policies and mechanisms affecting communication latency between virtual machines based on the Xen platform, and identify limitations that can result in long or unpredictable network stack latency for virtual machines deployed on this platform. To address these limitations, we propose and implement VATC , a Virtualization-Aware Traffic Control framework that supports differentiation (via rate-limited prioritization) of outbound and inbound network traffic from co-located virtualized hosts. Results of our experiments show how and why VATC can offer predictable (soft) latency guarantees to applications running on virtualized hosts with minimum overhead.
- 5 February 2021
- Published by Springer Nature in Real-Time Systems
- Vol. 57 (1-2) , 95-96
Guest Editorial: Special issue on outstanding papers from RTNS 2019
- 5 February 2021
- Published by Springer Nature in Real-Time Systems
- Vol. 57 (1-2) , 95-96
Guest Editorial: Special issue on outstanding papers from RTNS 2019
- 22 September 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-9
OpenCL Performance on the Intel Heterogeneous Architecture Research Platform
The fundamental operation of matrix multiplication is ubiquitous across a myriad of disciplines. Yet, the identification of new optimizations for matrix multiplication remains relevant for emerging hardware architectures and heterogeneous systems. Frameworks such as OpenCL enable computation orchestration on existing systems, and its availability using the Intel High Level Synthesis compiler allows users to architect new designs for reconfigurable hardware using C/C++. Using the HARPv2 as a vehicle for exploration, we investigate the utility of several traditional matrix multiplication optimizations to better understand the performance portability of OpenCL and the implications for such optimizations on cache coherent heterogeneous architectures. Our results give targeted insights into the applicability of best practices that were designed for existing architectures when used on emerging heterogeneous systems.
- 22 September 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-9
OpenCL Performance on the Intel Heterogeneous Architecture Research Platform
The fundamental operation of matrix multiplication is ubiquitous across a myriad of disciplines. Yet, the identification of new optimizations for matrix multiplication remains relevant for emerging hardware architectures and heterogeneous systems. Frameworks such as OpenCL enable computation orchestration on existing systems, and its availability using the Intel High Level Synthesis compiler allows users to architect new designs for reconfigurable hardware using C/C++. Using the HARPv2 as a vehicle for exploration, we investigate the utility of several traditional matrix multiplication optimizations to better understand the performance portability of OpenCL and the implications for such optimizations on cache coherent heterogeneous architectures. Our results give targeted insights into the applicability of best practices that were designed for existing architectures when used on emerging heterogeneous systems.
- 27 January 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 39 (12) , 5142-5155
Voltage-Stacked Power Delivery Systems: Reliability, Efficiency, and Power Management
In today’s manycore processors, energy loss of more than 20% may result from inherent inefficiencies of conventional power delivery system (PDS) design. By stacking multiple voltage domains in series to lower the step-down conversion ratio of the off-chip voltage regulator module (VRM) and reduce energy loss along the path of the power delivery network (PDN), voltage stacking (VS) offers a novel alternative power delivery technique to fundamentally improve power delivery efficiency (PDE). However, voltage stacking suffers from aggravated supply voltage noise from current imbalance, which hinders its adoption. In this paper, we investigate practical voltage stacking implementation in manycore processors to improve power delivery efficiency (PDE) and achieve reliable performance, while maintaining compatibility with advanced power management techniques. We first present the system configuration of a voltage-stacked manycore processor. We then systematically characterize supply voltage noise in voltage stacking, identify global and residual differential currents as its dominant contributors, and calculate the possible worst supply voltage noise. We next propose a hybrid voltage regulation solution, based on a charge-recycling off-chip voltage regulator and distributed integrated voltage regulators, to mitigate supply voltage noise effectively. We also study the compatibility of voltage stacking with higher level power management techniques. Finally, the performance of a voltage-stacked GPU system is comprehensively evaluated. Simulation results show that our approach can achieve 93.5% power delivery efficiency, reducing the power loss by 13.6% compared to conventional single-layer power delivery system.
- 27 January 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 39 (12) , 5142-5155
Voltage-Stacked Power Delivery Systems: Reliability, Efficiency, and Power Management
In today’s manycore processors, energy loss of more than 20% may result from inherent inefficiencies of conventional power delivery system (PDS) design. By stacking multiple voltage domains in series to lower the step-down conversion ratio of the off-chip voltage regulator module (VRM) and reduce energy loss along the path of the power delivery network (PDN), voltage stacking (VS) offers a novel alternative power delivery technique to fundamentally improve power delivery efficiency (PDE). However, voltage stacking suffers from aggravated supply voltage noise from current imbalance, which hinders its adoption. In this paper, we investigate practical voltage stacking implementation in manycore processors to improve power delivery efficiency (PDE) and achieve reliable performance, while maintaining compatibility with advanced power management techniques. We first present the system configuration of a voltage-stacked manycore processor. We then systematically characterize supply voltage noise in voltage stacking, identify global and residual differential currents as its dominant contributors, and calculate the possible worst supply voltage noise. We next propose a hybrid voltage regulation solution, based on a charge-recycling off-chip voltage regulator and distributed integrated voltage regulators, to mitigate supply voltage noise effectively. We also study the compatibility of voltage stacking with higher level power management techniques. Finally, the performance of a voltage-stacked GPU system is comprehensively evaluated. Simulation results show that our approach can achieve 93.5% power delivery efficiency, reducing the power loss by 13.6% compared to conventional single-layer power delivery system.
- 1 June 2017
- Published by Springer Nature in Real-Time Systems
- Vol. 53 (5) , 760-811
Mixed-criticality federated scheduling for parallel real-time tasks
- 1 June 2017
- Published by Springer Nature in Real-Time Systems
- Vol. 53 (5) , 760-811
Mixed-criticality federated scheduling for parallel real-time tasks