Model-Free Runtime Management of Concurrent Workloads for Energy-Efficient Many-Core Heterogeneous Systems

Aalsaud, Ali; Rafiev, Ashur; Xia, Fei; Shafik, Rishad; Yakovlev, Alex

doi:10.1109/patmos.2018.8464142

Cited by 6 publications

(14 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the most important questions for models is what they shall be used for. One use of a speedup model is in the optimisation of system operations via design‐time and run‐time management [8, 15, 30, 34, 36, 68, 101, 103–107].…”

Section: Discussionmentioning

confidence: 99%

“…The scheduling priority passes from high performance to low power cores in response to parallelism changes and a constant priority per core list cannot be built. This type of scheduling is especially relevant when trying to maximise performance without exceeding some power budget [105]. Two methods for generalising scheduling models were described in [38] to cover these kinds of characteristics.…”

Section: Parallelism and P‐fractions!mentioning

confidence: 99%

See 1 more Smart Citation

Amdahl's law in the context of heterogeneous many‐core systems – a survey

Al-hayanni

Xia

Rafiev

et al. 2020

IET Computers & Digital Techniques

Self Cite

View full text Add to dashboard Cite

For over 50 years, Amdahl's Law has been the hallmark model for reasoning about performance bounds for homogeneous parallel computing resources. As heterogeneous, many-core parallel resources continue to permeate into the modern server and embedded domains, there has been growing interest in promulgating realistic extensions and assumptions in keeping with newer use cases. This study aims to provide a comprehensive review of the purviews and insights provided by the extensive body of work related to Amdahl's law to date, focusing on computation speedup. The authors show that a significant portion of these studies has looked into analysing the scalability of the model considering both workload and system heterogeneity in real-world applications. The focus has been to improve the definition and semantic power of the two key parameters in the original model: the parallel fraction (f) and the computation capability improvement index (n). More recently, researchers have shown normal-form and multi-fraction extensions that can account for wider ranges of heterogeneity, validated on many-core systems running realistic workloads. Speedup models from Amdahl's law onwards have seen a wide range of uses, such as the optimisation of system execution, and these uses are even more important with the advent of the heterogeneous many-core era.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Parallelism and P‐fractions!mentioning

confidence: 99%

Amdahl's law in the context of heterogeneous many‐core systems – a survey

Al-hayanni

Xia

Rafiev

et al. 2020

IET Computers & Digital Techniques

Self Cite

View full text Add to dashboard Cite

show abstract

“…At an application level, different applications running concurrently lead to significant diversity in workload characteristics at run-time. At the hardware level, different actuation knobs such as degree of parallelism (DoP), dynamic voltage-frequency scaling (DVFS), and the type of active cores among a heterogeneous set expose a wide range of performance-energy trade-offs [1], [11], [15], [24]. Both these put together exacerbates the challenge of understanding application requirements, followed by allocating and scaling system resources to co-optimize performance and energy efficiency.…”

Section: Introductionmentioning

confidence: 99%

“…Despite the abundance of resource configuration choices to minimize energy consumption, the variation and diversity among different applications make it an exhaustive exploration [10], [18]. Further, on-line selection of resource configuration becomes even harder, considering an unknown sequence of concurrent applications, workload variation, inter-application interference and resource contention [1], [22]. Further, on-line selection of resource configuration becomes even harder, considering an unknown sequence of concurrent applications, workload variation, inter-application interference, and resource contention [1], [22].…”

Section: Introductionmentioning

confidence: 99%

“…Existing resource management approaches use offline profiling to characterize the applications at design time and use this information at run-time to avoid exhaustive search [10], [16], [21]. Other approaches use on-line prediction strategies, their efficiency and accuracy depend on the amount of power-performance statistics collected over a significant period of execution time [1], [18]. Both off-line and on-line approaches are confined to extract an optimal resource configuration of a single application running in isolation [3], [10], [12], [18], and are thus not readily adaptable to multiple concurrent application scenarios.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Concurrent Application Bias Scheduling for Energy Efficiency of Heterogeneous Multi-Core Platforms

Shamsa

Kanduri

Liljeberg

et al. 2022

IEEE Trans. Comput.

View full text Add to dashboard Cite

Minimizing energy consumption of concurrent applications on heterogeneous multi-core platforms is challenging given the diversity in energy-performance profiles of both the applications and hardware. Adaptive learning techniques made the exhaustive Pareto-optimal space exploration practically feasible to identify an energy efficient configuration. Existing approaches consider a single application's characteristic for optimizing energy consumption. However, an optimal configuration for a given application in isolation may not be optimal when other applications are run concurrently. Approaches that consider concurrent application scenarios overlook the weight of total energy consumption per application, restricting them from prioritizing among applications. We address this limitation by considering the mutual effect of concurrent applications on system wide energy consumption to adapt resource configuration at run-time. We characterize each application's power-performance profile as a weighted bias through off-line profiling. We infer this model combined with an on-line predictive strategy to make resource allocation decisions for minimizing energy consumption while honoring performance requirements. The proposed strategy is implemented as a user-space process and evaluated on a heterogeneous hardware platform of Odroid XU3 over the Rodinia benchmark suite. Experimental results show up to 61% of energy saving compared to the standard baseline of Linux governors and up to 27% of energy gain compared to state-of-the-art adaptive learning-based resource management techniques.

show abstract

A novel energy-efficient scheduling model for multi-core systems

Kumar

Vidyarthi

2020

Cluster Comput

View full text Add to dashboard Cite

Model-Free Runtime Management of Concurrent Workloads for Energy-Efficient Many-Core Heterogeneous Systems

Cited by 6 publications

References 17 publications

Amdahl's law in the context of heterogeneous many‐core systems – a survey

Amdahl's law in the context of heterogeneous many‐core systems – a survey

Concurrent Application Bias Scheduling for Energy Efficiency of Heterogeneous Multi-Core Platforms

A novel energy-efficient scheduling model for multi-core systems

Contact Info

Product

Resources

About