Pub Date : 2023-10-19DOI: 10.1109/TSUSC.2023.3325881
Akram Alofi;Mahmoud A. Bokhari;Rami Bahsoon;Robert Hendley
Blockchain technology has been widely adopted in many areas to provide more dependable and trustworthy systems, including digital infrastructure. Nevertheless, its widespread implementation is accompanied by significant environmental concerns, as it is considered a substantial contributor to greenhouse gas emissions. This environmental impact is mainly attributed to the inherent inefficiencies of its consensus algorithms, notably Proof of Work, which demands substantial computational power for trust establishment. This paper proposes a novel self-adaptive model to optimize the environmental sustainability of blockchain-based systems, addressing energy consumption and carbon emission without compromising the fundamental properties of blockchain technology. The model continuously monitors a blockchain-based system and adaptively selects miners, considering context changes and user needs. It dynamically selects a subset of miners to perform sustainable mining processes while ensuring the decentralization and trustworthiness of the system. The aim is to minimize blockchain-based systems' energy consumption and carbon emissions while maximizing their decentralization and trustworthiness. We conduct experiments to evaluate the efficiency and effectiveness of the model. The results show that our self-optimizing model can reduce energy consumption by 55.49% and carbon emissions by 71.25% on average while maintaining desirable levels of decentralization and trustworthiness by more than 96.08% and 75.12%, respectively. Furthermore, these enhancements can be achieved under different operating conditions compared to similar models, including the straightforward use of Proof of Work. Also, we have investigated and discussed the correlation between these objectives and how they are related to the number of miners within the blockchain-based systems.
{"title":"Self-Optimizing the Environmental Sustainability of Blockchain-Based Systems","authors":"Akram Alofi;Mahmoud A. Bokhari;Rami Bahsoon;Robert Hendley","doi":"10.1109/TSUSC.2023.3325881","DOIUrl":"10.1109/TSUSC.2023.3325881","url":null,"abstract":"Blockchain technology has been widely adopted in many areas to provide more dependable and trustworthy systems, including digital infrastructure. Nevertheless, its widespread implementation is accompanied by significant environmental concerns, as it is considered a substantial contributor to greenhouse gas emissions. This environmental impact is mainly attributed to the inherent inefficiencies of its consensus algorithms, notably Proof of Work, which demands substantial computational power for trust establishment. This paper proposes a novel self-adaptive model to optimize the environmental sustainability of blockchain-based systems, addressing energy consumption and carbon emission without compromising the fundamental properties of blockchain technology. The model continuously monitors a blockchain-based system and adaptively selects miners, considering context changes and user needs. It dynamically selects a subset of miners to perform sustainable mining processes while ensuring the decentralization and trustworthiness of the system. The aim is to minimize blockchain-based systems' energy consumption and carbon emissions while maximizing their decentralization and trustworthiness. We conduct experiments to evaluate the efficiency and effectiveness of the model. The results show that our self-optimizing model can reduce energy consumption by 55.49% and carbon emissions by 71.25% on average while maintaining desirable levels of decentralization and trustworthiness by more than 96.08% and 75.12%, respectively. Furthermore, these enhancements can be achieved under different operating conditions compared to similar models, including the straightforward use of Proof of Work. Also, we have investigated and discussed the correlation between these objectives and how they are related to the number of miners within the blockchain-based systems.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 3","pages":"396-408"},"PeriodicalIF":3.9,"publicationDate":"2023-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135058244","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-17DOI: 10.1109/TSUSC.2023.3325237
Yang Yang;Xuxun Liu;Kun Tang;Wenquan Che;Quan Xue
Charging scheduling plays a crucial role in ensuring durable operation for wireless rechargeable sensor networks. However, previous methods cannot meet the strict requirements of a high node survival rate and high energy usage effectiveness. In this article, we propose a multi-type charging scheduling strategy to meet such demands. In this strategy, the network is divided into an inner ring and an outer ring to satisfy different demands in different areas. The inner ring forms a flat topology, and adopts a periodic and single-node charging pattern mainly for a high node survival rate. A space priority and a time priority are designed to determine the charging sequence of the nodes. The optimal charging cycle and the optimal charging time are achieved by mathematical derivations. The outer ring forms a cluster topology, and adopts an on-demand and multi-node charging pattern mainly for high energy usage effectiveness. A space balancing principle and a time balancing principle are designed to determine the charging positions of the clusters. A gravitational search algorithm is designed to determine the charging sequence of the clusters. Several simulations verify the advantages of the proposed solution in terms of energy usage effectiveness, charging failure rate, and average task delay.
{"title":"Multi-Type Charging Scheduling Based on Area Requirement Difference for Wireless Rechargeable Sensor Networks","authors":"Yang Yang;Xuxun Liu;Kun Tang;Wenquan Che;Quan Xue","doi":"10.1109/TSUSC.2023.3325237","DOIUrl":"10.1109/TSUSC.2023.3325237","url":null,"abstract":"Charging scheduling plays a crucial role in ensuring durable operation for wireless rechargeable sensor networks. However, previous methods cannot meet the strict requirements of a high node survival rate and high energy usage effectiveness. In this article, we propose a multi-type charging scheduling strategy to meet such demands. In this strategy, the network is divided into an inner ring and an outer ring to satisfy different demands in different areas. The inner ring forms a flat topology, and adopts a periodic and single-node charging pattern mainly for a high node survival rate. A space priority and a time priority are designed to determine the charging sequence of the nodes. The optimal charging cycle and the optimal charging time are achieved by mathematical derivations. The outer ring forms a cluster topology, and adopts an on-demand and multi-node charging pattern mainly for high energy usage effectiveness. A space balancing principle and a time balancing principle are designed to determine the charging positions of the clusters. A gravitational search algorithm is designed to determine the charging sequence of the clusters. Several simulations verify the advantages of the proposed solution in terms of energy usage effectiveness, charging failure rate, and average task delay.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 2","pages":"182-196"},"PeriodicalIF":3.9,"publicationDate":"2023-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135007433","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-17DOI: 10.1109/TSUSC.2023.3325268
Renfeng Xiao;Xing He;Tingwen Huang;Junzhi Yu
In this paper, a novel Field-Programmable-Gate-Array (FPGA) implementation framework based on Lagrange programming neural network (LPNN), projection neural network (PNN) and proximal projection neural network (PPNN) is proposed which can be used to solve smooth and nonsmooth optimization problems. First, Count Unit (CU) and Calculate Unit (CaU) are designed for smooth problems with equality constraints, and these units are used to simulate the iteration actions of neural network (NN) and form a feedback loop with other basic digital circuit operations. Then, the optimal solutions of optimization problems are mapped by the output waveforms. Second, the digital circuit structures of Path Select Unit (PSU), projection operator and proximal operator are further designed to process the box constraints and nonsmooth terms, respectively. Finally, the effectiveness and feasibility of the circuit are verified by three numerical examples on the Quartus II 13.0 sp1 platform with the Cyclone IV E series chip EP4CE10F17C8.
本文提出了一种基于拉格朗日编程神经网络（LPNN）、投影神经网络（PNN）和近端投影神经网络（PPNN）的新型现场可编程门阵列（FPGA）实现框架，可用于解决平滑和非平滑优化问题。首先，针对具有相等约束条件的平滑问题设计了计数单元（CU）和计算单元（CaU），这些单元用于模拟神经网络（NN）的迭代动作，并与其他基本数字电路操作形成反馈回路。然后，通过输出波形映射出优化问题的最优解。其次，进一步设计了路径选择单元（PSU）、投影算子和近似算子的数字电路结构，以分别处理盒式约束和非光滑项。最后，在使用 Cyclone IV E 系列芯片 EP4CE10F17C8 的 Quartus II 13.0 sp1 平台上，通过三个数值实例验证了电路的有效性和可行性。
{"title":"FPGA Implementation of Classical Dynamic Neural Networks for Smooth and Nonsmooth Optimization Problems","authors":"Renfeng Xiao;Xing He;Tingwen Huang;Junzhi Yu","doi":"10.1109/TSUSC.2023.3325268","DOIUrl":"10.1109/TSUSC.2023.3325268","url":null,"abstract":"In this paper, a novel Field-Programmable-Gate-Array (FPGA) implementation framework based on Lagrange programming neural network (LPNN), projection neural network (PNN) and proximal projection neural network (PPNN) is proposed which can be used to solve smooth and nonsmooth optimization problems. First, Count Unit (CU) and Calculate Unit (CaU) are designed for smooth problems with equality constraints, and these units are used to simulate the iteration actions of neural network (NN) and form a feedback loop with other basic digital circuit operations. Then, the optimal solutions of optimization problems are mapped by the output waveforms. Second, the digital circuit structures of Path Select Unit (PSU), projection operator and proximal operator are further designed to process the box constraints and nonsmooth terms, respectively. Finally, the effectiveness and feasibility of the circuit are verified by three numerical examples on the Quartus II 13.0 sp1 platform with the Cyclone IV E series chip EP4CE10F17C8.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 2","pages":"197-208"},"PeriodicalIF":3.9,"publicationDate":"2023-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135002398","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-13DOI: 10.1109/TSUSC.2023.3324339
Shiya Liu;Yibin Liang;Yang Yi
Performing symbol detection for multiple-input and multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) systems is challenging and resource-consuming. In this paper, we present a liquid state machine (LSM), a type of reservoir computing based on spiking neural networks (SNNs), to achieve energy-efficient and sustainable symbol detection on the Loihi chip for MIMO-OFDM systems. SNNs are more biological-plausible and energy-efficient than conventional deep neural networks (DNN) but have lower performance in terms of accuracy. To enhance the accuracy of SNNs, we propose a knowledge distillation training algorithm called DNN-SNN co-learning, which employs a bi-directional learning path between a DNN and an SNN. Specifically, the knowledge from the output and intermediate layer of the DNN is transferred to the SNN, and we exploit a decoder to convert the spikes in the intermediate layers of an SNN into real numbers to enable communication between the DNN and the SNN. Through the bi-directional learning path, the SNN can mimic the behavior of the DNN by learning the knowledge from the DNN. Conversely, the DNN can better adapt itself to the SNN by using the knowledge from the SNN. We introduce a new loss function to enable knowledge distillation on regression tasks. Our LSM is implemented on Intel's Loihi neuromorphic chip, a specialized hardware platform for SNN models. The experimental results on symbol detection in MIMO-OFDM systems demonstrate that our LSM on the Loihi chip is more precise than conventional symbol detection algorithms. Also, the model consumes approximately 6 times less energy per sample than other quantized DNN-based models with comparable accuracy.
{"title":"DNN-SNN Co-Learning for Sustainable Symbol Detection in 5G Systems on Loihi Chip","authors":"Shiya Liu;Yibin Liang;Yang Yi","doi":"10.1109/TSUSC.2023.3324339","DOIUrl":"10.1109/TSUSC.2023.3324339","url":null,"abstract":"Performing symbol detection for multiple-input and multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) systems is challenging and resource-consuming. In this paper, we present a liquid state machine (LSM), a type of reservoir computing based on spiking neural networks (SNNs), to achieve energy-efficient and sustainable symbol detection on the Loihi chip for MIMO-OFDM systems. SNNs are more biological-plausible and energy-efficient than conventional deep neural networks (DNN) but have lower performance in terms of accuracy. To enhance the accuracy of SNNs, we propose a knowledge distillation training algorithm called DNN-SNN co-learning, which employs a bi-directional learning path between a DNN and an SNN. Specifically, the knowledge from the output and intermediate layer of the DNN is transferred to the SNN, and we exploit a decoder to convert the spikes in the intermediate layers of an SNN into real numbers to enable communication between the DNN and the SNN. Through the bi-directional learning path, the SNN can mimic the behavior of the DNN by learning the knowledge from the DNN. Conversely, the DNN can better adapt itself to the SNN by using the knowledge from the SNN. We introduce a new loss function to enable knowledge distillation on regression tasks. Our LSM is implemented on Intel's Loihi neuromorphic chip, a specialized hardware platform for SNN models. The experimental results on symbol detection in MIMO-OFDM systems demonstrate that our LSM on the Loihi chip is more precise than conventional symbol detection algorithms. Also, the model consumes approximately 6 times less energy per sample than other quantized DNN-based models with comparable accuracy.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 2","pages":"170-181"},"PeriodicalIF":3.9,"publicationDate":"2023-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136303279","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-13DOI: 10.1109/TSUSC.2023.3314916
Farui Wang;Meng Hao;Weizhe Zhang;Zheng Wang
GPUs play a central and indispensable role as accelerators in modern high-performance computing (HPC) platforms, enabling a wide range of tasks to be performed efficiently. However, the use of GPUs also results in significant energy consumption and carbon dioxide (CO2) emissions. This article presents MF-GPOEO, a model-free GPU online energy efficiency optimization framework. MF-GPOEO leverages a synthetic performance index and a PID controller to dynamically determine the optimal clock frequency configuration for GPUs. It profiles GPU kernel activity information under different frequency configurations and then compares GPU kernel execution time and gap duration between kernels to derive the synthetic performance index. With the performance index and measured average power, MF-GPOEO can use the PID controller to try different frequency configurations and find the optimal frequency configuration under the guidance of user-defined objective functions. We evaluate the MF-GPOEO by running it with 74 applications on an NVIDIA RTX3080Ti GPU. MF-GPOEO delivers a mean energy saving of 26.2% with a slight average execution time increase of 3.4% compared with NVIDIA's default clock scheduling strategy.
{"title":"Model-Free GPU Online Energy Optimization","authors":"Farui Wang;Meng Hao;Weizhe Zhang;Zheng Wang","doi":"10.1109/TSUSC.2023.3314916","DOIUrl":"10.1109/TSUSC.2023.3314916","url":null,"abstract":"GPUs play a central and indispensable role as accelerators in modern high-performance computing (HPC) platforms, enabling a wide range of tasks to be performed efficiently. However, the use of GPUs also results in significant energy consumption and carbon dioxide (CO2) emissions. This article presents MF-GPOEO, a model-free GPU online energy efficiency optimization framework. MF-GPOEO leverages a synthetic performance index and a PID controller to dynamically determine the optimal clock frequency configuration for GPUs. It profiles GPU kernel activity information under different frequency configurations and then compares GPU kernel execution time and gap duration between kernels to derive the synthetic performance index. With the performance index and measured average power, MF-GPOEO can use the PID controller to try different frequency configurations and find the optimal frequency configuration under the guidance of user-defined objective functions. We evaluate the MF-GPOEO by running it with 74 applications on an NVIDIA RTX3080Ti GPU. MF-GPOEO delivers a mean energy saving of 26.2% with a slight average execution time increase of 3.4% compared with NVIDIA's default clock scheduling strategy.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 2","pages":"141-154"},"PeriodicalIF":3.9,"publicationDate":"2023-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135402321","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-12DOI: 10.1109/TSUSC.2023.3314759
Longxin Zhang;Minghui Ai;Ke Liu;Jianguo Chen;Kenli Li
As the demand for Big Data analysis and artificial intelligence technology continues to surge, a significant amount of research has been conducted on cloud computing services. An effective workflow scheduling strategy stands as the pivotal factor in ensuring the quality of cloud services. Dynamic voltage and frequency scaling (DVFS) is an effective energy-saving technology that is extensively used in the development of workflow scheduling algorithms. However, DVFS reduces the processor's running frequency, which increases the possibility of soft errors in workflow execution, thereby lowering the workflow execution reliability. This study proposes an energy-aware reliability enhancement scheduling (EARES) method with a checkpoint mechanism to improve system reliability while meeting the workflow deadline and the energy consumption constraints. The proposed EARES algorithm consists of three phases, namely, workflow application initialization, deadline partitioning, and energy partitioning and virtual machine selection. Numerous experiments are conducted to assess the performance of the EARES algorithm using three real-world scientific workflows. Experimental results demonstrate that the EARES algorithm remarkably improves reliability in comparison with other state-of-the-art algorithms while meeting the deadline and satisfying the energy consumption requirement.
{"title":"Reliability Enhancement Strategies for Workflow Scheduling Under Energy Consumption Constraints in Clouds","authors":"Longxin Zhang;Minghui Ai;Ke Liu;Jianguo Chen;Kenli Li","doi":"10.1109/TSUSC.2023.3314759","DOIUrl":"10.1109/TSUSC.2023.3314759","url":null,"abstract":"As the demand for Big Data analysis and artificial intelligence technology continues to surge, a significant amount of research has been conducted on cloud computing services. An effective workflow scheduling strategy stands as the pivotal factor in ensuring the quality of cloud services. Dynamic voltage and frequency scaling (DVFS) is an effective energy-saving technology that is extensively used in the development of workflow scheduling algorithms. However, DVFS reduces the processor's running frequency, which increases the possibility of soft errors in workflow execution, thereby lowering the workflow execution reliability. This study proposes an energy-aware reliability enhancement scheduling (EARES) method with a checkpoint mechanism to improve system reliability while meeting the workflow deadline and the energy consumption constraints. The proposed EARES algorithm consists of three phases, namely, workflow application initialization, deadline partitioning, and energy partitioning and virtual machine selection. Numerous experiments are conducted to assess the performance of the EARES algorithm using three real-world scientific workflows. Experimental results demonstrate that the EARES algorithm remarkably improves reliability in comparison with other state-of-the-art algorithms while meeting the deadline and satisfying the energy consumption requirement.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 2","pages":"155-169"},"PeriodicalIF":3.9,"publicationDate":"2023-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135401489","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-11DOI: 10.1109/TSUSC.2023.3313770
Zhao Tong;Jinhui Cai;Jing Mei;Kenli Li;Keqin Li
The efficiency of production and equipment maintenance costs in the Industrial Internet of Things (IIoT) are directly impacted by equipment lifetime, making it an important concern. Mobile edge computing (MEC) can enhance network performance, extend device lifetime, and effectively reduce carbon emissions by integrating energy harvesting (EH) technology. However, when the two are combined, the coupling effect of energy and the system's communication resource management pose a great challenge to the development of computational offloading strategies. This paper investigates the problem of maximizing the energy efficiency of computation offloading in a two-tier MEC network powered by wireless power transfer (WPT). First, the corresponding mathematical models are developed for local computing, edge server processing, communication, and EH. The proposed fractional problem is transformed into a stochastic optimization problem by Dinkelbach method. In addition, virtual power queues are introduced to eliminate energy coupling effects by maintaining the stability of the battery power queues. Next, the problem is then resolved through the utilization of both Lyapunov optimization and convex optimization method. Consequently, a wireless energy transmission-based algorithm for maximizing energy efficiency is proposed. Finally, energy efficiency, an important parameter of network performance, is used as an indicator. The excellent performance of the EEMA-WET algorithm is verified through extensive extension and comparison experiments.
{"title":"Computation Offloading for Energy Efficiency Maximization of Sustainable Energy Supply Network in IIoT","authors":"Zhao Tong;Jinhui Cai;Jing Mei;Kenli Li;Keqin Li","doi":"10.1109/TSUSC.2023.3313770","DOIUrl":"10.1109/TSUSC.2023.3313770","url":null,"abstract":"The efficiency of production and equipment maintenance costs in the Industrial Internet of Things (IIoT) are directly impacted by equipment lifetime, making it an important concern. Mobile edge computing (MEC) can enhance network performance, extend device lifetime, and effectively reduce carbon emissions by integrating energy harvesting (EH) technology. However, when the two are combined, the coupling effect of energy and the system's communication resource management pose a great challenge to the development of computational offloading strategies. This paper investigates the problem of maximizing the energy efficiency of computation offloading in a two-tier MEC network powered by wireless power transfer (WPT). First, the corresponding mathematical models are developed for local computing, edge server processing, communication, and EH. The proposed fractional problem is transformed into a stochastic optimization problem by Dinkelbach method. In addition, virtual power queues are introduced to eliminate energy coupling effects by maintaining the stability of the battery power queues. Next, the problem is then resolved through the utilization of both Lyapunov optimization and convex optimization method. Consequently, a wireless energy transmission-based algorithm for maximizing energy efficiency is proposed. Finally, energy efficiency, an important parameter of network performance, is used as an indicator. The excellent performance of the EEMA-WET algorithm is verified through extensive extension and comparison experiments.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 2","pages":"128-140"},"PeriodicalIF":3.9,"publicationDate":"2023-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135361324","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-11DOI: 10.1109/TSUSC.2023.3313880
Ke Wang;Hao Zheng;Jiajun Li;Ahmed Louri
While current Graph Convolutional Networks (GCNs) accelerators have achieved notable success in a wide range of application domains, these GCN accelerators can not support various intra- and inter- GCN dataflows or adapt to diverse GCN applications. In this paper, we propose Morph-GCNX, a flexible GCN accelerator architecture for high-performance and energy-efficient GCN execution. The proposed design consists of a flexible Processing Element (PE) array that can be partitioned at runtime and adapt to the computational needs of different layers within a GCN or multiple concurrent GCNs. The proposed Morph-GCNX also consists of a morphable interconnection design to support a wide range of GCN dataflows with various parallelization and data reuse strategies for GCN execution. We also propose a hardware-application co-exploration technique that explores the GCN and hardware design spaces to identify the best PE partition, workload allocation, dataflow, and interconnection configurations, with the goal of improving overall performance and energy. Simulation results show that the proposed Morph-GCNX architecture achieves 18.8×, 2.9×, 1.9×, 1.8×, and 2.5× better performance, reduces DRAM accesses by a factor of 10.8×, 3.7×, 2.2×, 2.5×, and 1.3×, and improves energy consumption by 13.2×, 5.6×, 2.1×, 2.5×, and 1.3×, as compared to prior designs including HyGCN, AWB-GCN, LW-GCN, GCoD, and GCNAX, respectively.
{"title":"Morph-GCNX: A Universal Architecture for High-Performance and Energy-Efficient Graph Convolutional Network Acceleration","authors":"Ke Wang;Hao Zheng;Jiajun Li;Ahmed Louri","doi":"10.1109/TSUSC.2023.3313880","DOIUrl":"10.1109/TSUSC.2023.3313880","url":null,"abstract":"While current Graph Convolutional Networks (GCNs) accelerators have achieved notable success in a wide range of application domains, these GCN accelerators can not support various intra- and inter- GCN dataflows or adapt to diverse GCN applications. In this paper, we propose Morph-GCNX, a flexible GCN accelerator architecture for high-performance and energy-efficient GCN execution. The proposed design consists of a flexible Processing Element (PE) array that can be partitioned at runtime and adapt to the computational needs of different layers within a GCN or multiple concurrent GCNs. The proposed Morph-GCNX also consists of a morphable interconnection design to support a wide range of GCN dataflows with various parallelization and data reuse strategies for GCN execution. We also propose a hardware-application co-exploration technique that explores the GCN and hardware design spaces to identify the best PE partition, workload allocation, dataflow, and interconnection configurations, with the goal of improving overall performance and energy. Simulation results show that the proposed Morph-GCNX architecture achieves 18.8×, 2.9×, 1.9×, 1.8×, and 2.5× better performance, reduces DRAM accesses by a factor of 10.8×, 3.7×, 2.2×, 2.5×, and 1.3×, and improves energy consumption by 13.2×, 5.6×, 2.1×, 2.5×, and 1.3×, as compared to prior designs including HyGCN, AWB-GCN, LW-GCN, GCoD, and GCNAX, respectively.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 2","pages":"115-127"},"PeriodicalIF":3.9,"publicationDate":"2023-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135361631","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-01DOI: 10.1109/TSUSC.2023.3303637
Xiaoyun Han;Chaoxu Mu;Jiebei Zhu;Hongjie Jia
With the increasing scale of cloud data centers (CDCs), the energy consumption of CDCs is sharply increasing. In this article, an efficient energy-saving strategy is proposed for CDCs. The greedy virtual machine (VM) deployment strategy is obtained by using the least number of servers, the heuristic VM migration strategy is obtained by using the improved double threshold algorithm, and the comprehensive VM scheduling strategy of severs is obtained by combining deployment and migration strategies. Furthermore, for the privacy security of VM scheduling, a safety-oriented energy-saving scheme based on information difference is proposed to ensure the dataset availability under privacy protection, comparing with $varepsilon$ -differential privacy algorithm and $(varepsilon, delta)$ -differential privacy algorithm. Simulation results show that the safe energy-saving strategy can significantly reduce the energy consumption in CDCs with guaranteeing the security and availability of the important datasets.
{"title":"A Safe Virtual Machine Scheduling Strategy for Energy Conservation and Privacy Protection of Server Clusters in Cloud Data Centers","authors":"Xiaoyun Han;Chaoxu Mu;Jiebei Zhu;Hongjie Jia","doi":"10.1109/TSUSC.2023.3303637","DOIUrl":"10.1109/TSUSC.2023.3303637","url":null,"abstract":"With the increasing scale of cloud data centers (CDCs), the energy consumption of CDCs is sharply increasing. In this article, an efficient energy-saving strategy is proposed for CDCs. The greedy virtual machine (VM) deployment strategy is obtained by using the least number of servers, the heuristic VM migration strategy is obtained by using the improved double threshold algorithm, and the comprehensive VM scheduling strategy of severs is obtained by combining deployment and migration strategies. Furthermore, for the privacy security of VM scheduling, a safety-oriented energy-saving scheme based on information difference is proposed to ensure the dataset availability under privacy protection, comparing with \u0000<inline-formula><tex-math>$varepsilon$</tex-math></inline-formula>\u0000-differential privacy algorithm and \u0000<inline-formula><tex-math>$(varepsilon, delta)$</tex-math></inline-formula>\u0000-differential privacy algorithm. Simulation results show that the safe energy-saving strategy can significantly reduce the energy consumption in CDCs with guaranteeing the security and availability of the important datasets.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 1","pages":"46-60"},"PeriodicalIF":3.9,"publicationDate":"2023-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83911053","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In recent years, serverless computing has garnered significant attention owing to its high scalability, pay-as-you-go billing model, and efficient resource management provided by cloud service providers. Optimal resource scheduling of serverless computing has become imperative to reduce energy consumption and enable sustainable computing. However, existing serverless platforms encounter two significant challenges: the cold-start problem of containers and the absence of an effective resource allocation strategy for serverless workflows. Existing pre-warm strategies are associated with high computational overhead, while current resource scheduling techniques inadequately account for the intricate structure of serverless workflows. To address these challenges, we present SSC, a pre-warming and automatic resource allocation framework designed explicitly for serverless workflows. We introduce an innovative gradient-based algorithm for pre-warming containers, significantly reducing cold start hit rates. Moreover, leveraging a critical path and priority queue-based algorithm, SSC enables efficient allocation of resources for serverless workflows. In our experimental evaluation, SSC reduces the cold start hit rate by nearly 50% and achieves substantial cost savings of approximately 30%.
{"title":"Sustainable Serverless Computing With Cold-Start Optimization and Automatic Workflow Resource Scheduling","authors":"Shanxing Pan;Hongyu Zhao;Zinuo Cai;Dongmei Li;Ruhui Ma;Haibing Guan","doi":"10.1109/TSUSC.2023.3311197","DOIUrl":"10.1109/TSUSC.2023.3311197","url":null,"abstract":"In recent years, serverless computing has garnered significant attention owing to its high scalability, pay-as-you-go billing model, and efficient resource management provided by cloud service providers. Optimal resource scheduling of serverless computing has become imperative to reduce energy consumption and enable sustainable computing. However, existing serverless platforms encounter two significant challenges: the cold-start problem of containers and the absence of an effective resource allocation strategy for serverless workflows. Existing pre-warm strategies are associated with high computational overhead, while current resource scheduling techniques inadequately account for the intricate structure of serverless workflows. To address these challenges, we present SSC, a pre-warming and automatic resource allocation framework designed explicitly for serverless workflows. We introduce an innovative gradient-based algorithm for pre-warming containers, significantly reducing cold start hit rates. Moreover, leveraging a critical path and priority queue-based algorithm, SSC enables efficient allocation of resources for serverless workflows. In our experimental evaluation, SSC reduces the cold start hit rate by nearly 50% and achieves substantial cost savings of approximately 30%.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 3","pages":"329-340"},"PeriodicalIF":3.9,"publicationDate":"2023-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72796524","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：