Using Multiple Input, Multiple Output Formal Control to Maximize Resource Efficiency in Architectures

Pothukuchi, Raghavendra Pradyumna; Ansari, Amin; Voulgaris, Petros G.; Torrellas, Josep

doi:10.1109/isca.2016.63

Cited by 41 publications

(29 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A typical approach to manage resources of a multi-core system is to independently control each component, possibly with a different objective. For example, in [14][15][16] the voltagefrequency (VF) of the processor is controlled to meet a specific performance target; while, on the other hand, the shared last level cache (LLC) is usually partitioned to minimize the total number of cache misses [7]. Nevertheless, in a system under per-application performance constraints, independent schemes are not very effective.…”

Section: Summary Of Paper Imentioning

confidence: 99%

QoS-Driven Coordinated Management of Resources to Save Energy in Multi-core Systems

Nejat

Pericàs

Stenström

2019

2019 IEEE International Parallel and Distributed Processing Symposium (IPDPS)

View full text Add to dashboard Cite

Reducing the energy consumption of computing systems is a necessary endeavor. However, saving energy should not come at the expense of degrading user experience. To this end, in this thesis, we assume that applications running on multi-core processors are associated with a quality-of-service (QoS) target in terms of performance constraints. This way, hardware resources can be throttled to minimize energy expenditure without violating the QoS requirements.Typical resource management schemes control different resources such as processor cores and on-chip cache memory independently. These approaches are not effective under performance constraints for all applications. Therefore, this thesis presents multi-core resource management schemes that coordinately control several resources in a unified algorithm. This way, the resource manger can find trade-offs between resource allocations to different applications to reduce system-level energy consumption, while still meeting the QoS targets expressed as performance constraints for every application.Implementing a coordinated resource management scheme that dynamically adapts to varying run time behavior of a multi-programmed workload without any prior knowledge about the applications is a challenging task. Two different schemes are presented in this thesis to address this challenge. Both schemes are invoked at regular intervals during program execution. They employ simple and, yet, sufficiently accurate analytical models and a novel hardware technique to predict the effect of different resource allocations on performance and energy for each application. Using a heuristic method, the multi-dimensional system configuration space is pruned in several levels to find the optimum resource settings, with respect to energy efficiency, in a negligible time.In the first scheme a resource management algorithm is presented that coordinates the control of voltage-frequency (VF) of each processor core with partitioning of the on-chip cache space. In the second scheme, a re-configurable processor is considered in which sections of the core micro-architectural resources can be dynamically deactivated to save energy. The resource manager can reactivate these sections, at the proper time, to increase instruction and memory level parallelism (ILP/MLP). This introduces new trade-offs between processor core size, VF settings, and the allocation of cache space for each application. By exploiting these trade-offs, the second scheme improves the energy savings compared to the first scheme considerably.The proposed schemes are evaluated using a novel simulation framework. This framework estimates the effect of different resource management algorithms on full execution of benchmark applications in a multi-programmed workload. According to the experimental results, the proposed schemes can save up to 18% of system energy while respecting the performance constraints of all applications. The average energy savings are 6% and 10% with the first and second schemes, respectively. Further experiments on ...

show abstract

Section: Summary Of Paper Imentioning

confidence: 99%

QoS-Driven Coordinated Management of Resources to Save Energy in Multi-core Systems

Nejat

Pericàs

Stenström

2019

2019 IEEE International Parallel and Distributed Processing Symposium (IPDPS)

View full text Add to dashboard Cite

show abstract

“…These techniques combine PID-like control techniques to various parts of the system, and provide empirical demonstrations of overall system behavior without taking domain knowledge into consideration. Recent work on optimal resource allocation focuses on designing sophisticated control systems using linear quadratic Gaussian control for example [42]- [44]. Unlike SLAMBooster, they also do not exploit application-specific properties to perform control.…”

Section: B Adaptive Resource Managementmentioning

confidence: 99%

SLAMBooster: An Application-Aware Online Controller for Approximation in Dense SLAM

Pei

Biswas

Fussell

et al. 2019

2019 28th International Conference on Parallel Architectures and Compilation Techniques (PACT)

View full text Add to dashboard Cite

Simultaneous Localization and Mapping (SLAM) is the problem of constructing a map of a mobile agent's environment while localizing the agent within the map. Dense SLAM algorithms perform reconstruction and localization at pixel granularity. These algorithms require a lot of computational power, which has hindered their use on low-power resourceconstrained devices.Approximate computing can be used to speed up SLAM implementations as long as the approximations do not prevent the agent from navigating correctly through the environment. Previous studies of approximation in SLAM have assumed that the entire trajectory of the agent is known before the agent starts, and they have focused on offline controllers that set approximation knobs at the start of the trajectory. In practice, the trajectory is usually not known ahead of time, and allowing knob settings to change dynamically opens up more opportunities for reducing computation time and energy.In this paper, we describe SLAMBooster, an application-aware, online control system for dense SLAM that adaptively controls approximation knobs during the motion of the agent. SLAMBooster is based on a control technique called proportional-integralderivative (PID) controller but our experiments showed this application-agnostic controller led to an unacceptable reduction in localization accuracy. To address this problem, SLAMBooster also exploits domain knowledge for controlling approximation by performing smooth surface detection and pose correction.We implemented SLAMBooster in the open-source SLAM-Bench framework and evaluated it on more than a dozen trajectories from both the literature and our own study. Our experiments show that on the average, SLAMBooster reduces the computation time by 72% and energy consumption by 35% on an embedded platform, while maintaining the accuracy of localization within reasonable bounds. These improvements make it feasible to deploy SLAM on a wider range of devices.

show abstract

“…Focusing on one type of resource may lead to insufficient CPU resources but excessive memory resources. In addition, some studies used SISO control to allocate resources by maximizing resource utilization [3][4][5]. However, maximizing resource utilization does not mean efficiently use of resources, as excessive resource utilization rate may lead to poor QoS.…”

Section: Related Workmentioning

confidence: 99%

“…Recent studies have focused on dynamic resource allocation by maximizing resource utilization in cloud computing systems [3][4][5]. Maximizing resource utilization provides many benefits for both users and service providers; for example, it can greatly reduce users' costs [6].…”

Section: Introductionmentioning

confidence: 99%

An Adaptive Control Method for Resource Provisioning with Resource Utilization Constraints in Cloud Computing

Gong¹,

Yin²,

Zhang³

et al. 2019

IJCIS

View full text Add to dashboard Cite

A B S T R A C TCloud computing enables users to purchase virtual resources on demand; therefore, the service requests change over time. Dynamic resource provisioning for cloud computing has become a key challenge. To reduce the associated costs, resource utilization must be improved, and ensure the Quality of Service (QoS) to meet the service-level agreements (SLAs). However, it is difficult to improve the resource utilization level and maintain a high QoS with fluctuating workloads in shared cloud computing systems. In this paper, we propose an adaptive control method for resource provisioning in cloud computing systems to simultaneously improve the resource utilization, achieve a satisfactory QoS, and react to the dynamic workload. We proposed an approach integrating adaptive multi-input and multi-output (MIMO) control and radial basis function (RBF) neural network to react to the highly dynamic workloads, and precisely control the resource allocation to improve resource utilization based on the maximum allowable QoS requirements. Experiments based on real-world workloads show that the proposed approach jointly improves resource utilization, maintains a satisfactory QoS, and handles workload fluctuations in a coordinated manner.

show abstract

Using Multiple Input, Multiple Output Formal Control to Maximize Resource Efficiency in Architectures

Cited by 41 publications

References 47 publications

QoS-Driven Coordinated Management of Resources to Save Energy in Multi-core Systems

QoS-Driven Coordinated Management of Resources to Save Energy in Multi-core Systems

SLAMBooster: An Application-Aware Online Controller for Approximation in Dense SLAM

An Adaptive Control Method for Resource Provisioning with Resource Utilization Constraints in Cloud Computing

Contact Info

Product

Resources

About