Pub Date : 2023-01-01DOI: 10.34133/icomputing.0041
Chenyang Wu, Zongzhang Zhang
Reinforcement learning (RL) is indispensable for building intelligent decision-making agents. However, current RL algorithms suffer from statistical and computational inefficiencies that render them useless in most real-world applications. We argue that high-value information in the real world is essential for intelligent decision-making; however, it is not addressed by most RL formalisms. Through a closer investigation of high-value information, it becomes evident that, to exploit high-value information, there is a need to formalize intelligent decision-making as bounded-optimal lifelong RL. Thus, the challenge of achieving intelligent decision-making is summarized as effectively surfing information, specifically regarding handling the non-IID (independent and identically distributed) information stream while operating with limited resources. This study discusses the design of an intelligent decision-making agent and examines its primary challenges, which are (a) online learning for non-IID data streams, (b) efficient reasoning with limited resources, and (c) the exploration–exploitation dilemma. We review relevant problems and research in the field of RL literature and conclude that current RL methods are insufficient to address these challenges. We propose that an agent capable of overcoming these challenges could effectively surf the information overload in the real world and achieve sample- and compute-efficient intelligent decision-making.
{"title":"Surfing Information: The Challenge of Intelligent Decision-Making","authors":"Chenyang Wu, Zongzhang Zhang","doi":"10.34133/icomputing.0041","DOIUrl":"https://doi.org/10.34133/icomputing.0041","url":null,"abstract":"Reinforcement learning (RL) is indispensable for building intelligent decision-making agents. However, current RL algorithms suffer from statistical and computational inefficiencies that render them useless in most real-world applications. We argue that high-value information in the real world is essential for intelligent decision-making; however, it is not addressed by most RL formalisms. Through a closer investigation of high-value information, it becomes evident that, to exploit high-value information, there is a need to formalize intelligent decision-making as bounded-optimal lifelong RL. Thus, the challenge of achieving intelligent decision-making is summarized as effectively surfing information, specifically regarding handling the non-IID (independent and identically distributed) information stream while operating with limited resources. This study discusses the design of an intelligent decision-making agent and examines its primary challenges, which are (a) online learning for non-IID data streams, (b) efficient reasoning with limited resources, and (c) the exploration–exploitation dilemma. We review relevant problems and research in the field of RL literature and conclude that current RL methods are insufficient to address these challenges. We propose that an agent capable of overcoming these challenges could effectively surf the information overload in the real world and achieve sample- and compute-efficient intelligent decision-making.","PeriodicalId":45291,"journal":{"name":"International Journal of Intelligent Computing and Cybernetics","volume":"14 1","pages":""},"PeriodicalIF":4.3,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74764103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.34133/icomputing.0064
Philip N Johnson-Laird, Marco Ragni
Today, chatbots and other artificial intelligence tools pass the Turing test, which was Turing’s alternative to trying to answer the question: can a machine think? Despite their success in passing the Turing test, these machines do not think. We therefore propose a test of a more focused question: does a program reason in the way that humans reason? This test treats an “intelligent” program as though it were a participant in a psychological study and has 3 steps: (a) test the program in a set of experiments examining its inferences, (b) test its understanding of its own way of reasoning, and (c) examine, if possible, the cognitive adequacy of the source code for the program.
{"title":"What Should Replace the Turing Test?","authors":"Philip N Johnson-Laird, Marco Ragni","doi":"10.34133/icomputing.0064","DOIUrl":"https://doi.org/10.34133/icomputing.0064","url":null,"abstract":"Today, chatbots and other artificial intelligence tools pass the Turing test, which was Turing’s alternative to trying to answer the question: can a machine think? Despite their success in passing the Turing test, these machines do not think. We therefore propose a test of a more focused question: does a program reason in the way that humans reason? This test treats an “intelligent” program as though it were a participant in a psychological study and has 3 steps: (a) test the program in a set of experiments examining its inferences, (b) test its understanding of its own way of reasoning, and (c) examine, if possible, the cognitive adequacy of the source code for the program.","PeriodicalId":45291,"journal":{"name":"International Journal of Intelligent Computing and Cybernetics","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135051490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.34133/icomputing.0032
Aris Tsirigotis, G. Sarantoglou, M. Skontranis, S. Deligiannidis, Kostas Sozos, Giannis Tsilikas, Dimitris Dermanis, A. Bogris, C. Mesaritakis
We provide an overview of the rapidly evolving landscape of integrated photonic neuromorphic architectures, specifically targeting the implementation of convolutional neural networks. The exploding research momentum stems from the well-known advantages of photonic circuits compared to digital electronics, and at the same time, it is driven by the massive need for cognitive image/video processing. In this context, we provide a detailed literature review on photonic cores operating as convolutional neural networks, covering either the functionality of a conventional neural network or its spiking counterpart. Moreover, we propose 2 alternative photonic approaches that refrain from simply transferring neural network concepts directly into the optical domain; instead, they focus on fusing photonic, digital electronic, and event-based bioinspired processing to optimally exploit the virtues of each scheme. These approaches can offer beyond state-of-the-art performance while relying on realistic, scalable technology. The first approach is based on a photonic integrated platform and a bioinspired spectrum-slicing technique. The photonic chip allows feature extraction through optical filtering with low power consumption and an equivalent computational efficiency of 72 femtojoules per multiply-and-accumulate operation for 5-bit precision. When combined with typical digital neural networks, an almost 5-fold reduction in the number of parameters was achieved with a minor loss of accuracy compared to established convolutional neural networks. The second approach follows a bioisomorphic route in which miniaturized spiking laser neurons and unsupervised bioinspired training are unified in a deep architecture, revealing a noise-resilient and power-efficient proposition.
{"title":"Unconventional Integrated Photonic Accelerators for High-Throughput Convolutional Neural Networks","authors":"Aris Tsirigotis, G. Sarantoglou, M. Skontranis, S. Deligiannidis, Kostas Sozos, Giannis Tsilikas, Dimitris Dermanis, A. Bogris, C. Mesaritakis","doi":"10.34133/icomputing.0032","DOIUrl":"https://doi.org/10.34133/icomputing.0032","url":null,"abstract":"We provide an overview of the rapidly evolving landscape of integrated photonic neuromorphic architectures, specifically targeting the implementation of convolutional neural networks. The exploding research momentum stems from the well-known advantages of photonic circuits compared to digital electronics, and at the same time, it is driven by the massive need for cognitive image/video processing. In this context, we provide a detailed literature review on photonic cores operating as convolutional neural networks, covering either the functionality of a conventional neural network or its spiking counterpart. Moreover, we propose 2 alternative photonic approaches that refrain from simply transferring neural network concepts directly into the optical domain; instead, they focus on fusing photonic, digital electronic, and event-based bioinspired processing to optimally exploit the virtues of each scheme. These approaches can offer beyond state-of-the-art performance while relying on realistic, scalable technology. The first approach is based on a photonic integrated platform and a bioinspired spectrum-slicing technique. The photonic chip allows feature extraction through optical filtering with low power consumption and an equivalent computational efficiency of 72 femtojoules per multiply-and-accumulate operation for 5-bit precision. When combined with typical digital neural networks, an almost 5-fold reduction in the number of parameters was achieved with a minor loss of accuracy compared to established convolutional neural networks. The second approach follows a bioisomorphic route in which miniaturized spiking laser neurons and unsupervised bioinspired training are unified in a deep architecture, revealing a noise-resilient and power-efficient proposition.","PeriodicalId":45291,"journal":{"name":"International Journal of Intelligent Computing and Cybernetics","volume":"41 1","pages":""},"PeriodicalIF":4.3,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83221305","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.1007/978-3-031-37717-4
{"title":"Intelligent Computing: Proceedings of the 2023 Computing Conference, Volume 1","authors":"","doi":"10.1007/978-3-031-37717-4","DOIUrl":"https://doi.org/10.1007/978-3-031-37717-4","url":null,"abstract":"","PeriodicalId":45291,"journal":{"name":"International Journal of Intelligent Computing and Cybernetics","volume":"25 1","pages":""},"PeriodicalIF":4.3,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85965514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.1007/978-3-031-37963-5
{"title":"Intelligent Computing: Proceedings of the 2023 Computing Conference, Volume 2","authors":"","doi":"10.1007/978-3-031-37963-5","DOIUrl":"https://doi.org/10.1007/978-3-031-37963-5","url":null,"abstract":"","PeriodicalId":45291,"journal":{"name":"International Journal of Intelligent Computing and Cybernetics","volume":"24 1","pages":""},"PeriodicalIF":4.3,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90504699","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.34133/icomputing.0014
Gao Wang, Daniele Faccio
Brain–computer interfaces are enabling a range of new possibilities and routes for augmenting human capability. Here, we propose brain–computer interfaces as a route towards forms of computation, i.e., computational imaging, that blend the brain with external silicon processing. We demonstrate ghost imaging of a hidden scene using the human visual system that is combined with an adaptive computational imaging scheme. This is achieved through a projection pattern “carving” technique that relies on real-time feedback from the brain to modify patterns at the light projector, thus enabling more efficient and higher-resolution imaging. This brain–computer connectivity demonstrates a form of augmented human computation that could, in the future, extend the sensing range of human vision and provide new approaches to the study of the neurophysics of human perception. As an example, we illustrate a simple experiment whereby image reconstruction quality is affected by simultaneous conscious processing and readout of the perceived light intensities.
{"title":"Computational Ghost Imaging with the Human Brain","authors":"Gao Wang, Daniele Faccio","doi":"10.34133/icomputing.0014","DOIUrl":"https://doi.org/10.34133/icomputing.0014","url":null,"abstract":"Brain–computer interfaces are enabling a range of new possibilities and routes for augmenting human capability. Here, we propose brain–computer interfaces as a route towards forms of computation, i.e., computational imaging, that blend the brain with external silicon processing. We demonstrate ghost imaging of a hidden scene using the human visual system that is combined with an adaptive computational imaging scheme. This is achieved through a projection pattern “carving” technique that relies on real-time feedback from the brain to modify patterns at the light projector, thus enabling more efficient and higher-resolution imaging. This brain–computer connectivity demonstrates a form of augmented human computation that could, in the future, extend the sensing range of human vision and provide new approaches to the study of the neurophysics of human perception. As an example, we illustrate a simple experiment whereby image reconstruction quality is affected by simultaneous conscious processing and readout of the perceived light intensities.","PeriodicalId":45291,"journal":{"name":"International Journal of Intelligent Computing and Cybernetics","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134996518","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-01DOI: 10.34133/icomputing.0031
Dafydd Owen-Newns, Joshua Robertson, Matěj Hejda, Antonio Hurtado
Photonic technologies offer great prospects for novel, ultrafast, energy-efficient, and hardware-friendly neuromorphic (brain-like) computing platforms. Moreover, neuromorphic photonic approaches based on ubiquitous, technology-mature, and low-cost vertical-cavity surface-emitting lasers (VCSELs) (devices found in fiber-optic transmitters, mobile phones, and automotive sensors) are of particular interest. Given that VCSELs have shown the ability to realize neuronal optical spiking responses (at ultrafast GHz rates), their use in spike-based information-processing systems has been proposed. In this study, spiking neural network (SNN) operation, based on a hardware-friendly photonic system of just one VCSEL, is reported alongside a novel binary weight “significance” training scheme that fully capitalizes on the discrete nature of the optical spikes used by the SNN to process input information. The VCSEL-based photonic SNN was tested with a highly complex multivariate classification task (MADELON) before its performance was compared using a traditional least-squares training method and an alternative novel binary weighting scheme. Excellent classification accuracies of >94% were achieved by both training methods, exceeding the benchmark performance of the dataset in a fraction of the processing time. The newly reported training scheme also dramatically reduces the training set size requirements and the number of trained nodes (≤1% of the total network node count). This VCSEL-based photonic SNN, in combination with the reported “significance” weighting scheme, therefore grants ultrafast spike-based optical processing highly reduced training requirements and hardware complexity for potential application in future neuromorphic systems and artificial intelligence applications.
{"title":"Photonic Spiking Neural Networks with Highly Efficient Training Protocols for Ultrafast Neuromorphic Computing Systems","authors":"Dafydd Owen-Newns, Joshua Robertson, Matěj Hejda, Antonio Hurtado","doi":"10.34133/icomputing.0031","DOIUrl":"https://doi.org/10.34133/icomputing.0031","url":null,"abstract":"Photonic technologies offer great prospects for novel, ultrafast, energy-efficient, and hardware-friendly neuromorphic (brain-like) computing platforms. Moreover, neuromorphic photonic approaches based on ubiquitous, technology-mature, and low-cost vertical-cavity surface-emitting lasers (VCSELs) (devices found in fiber-optic transmitters, mobile phones, and automotive sensors) are of particular interest. Given that VCSELs have shown the ability to realize neuronal optical spiking responses (at ultrafast GHz rates), their use in spike-based information-processing systems has been proposed. In this study, spiking neural network (SNN) operation, based on a hardware-friendly photonic system of just one VCSEL, is reported alongside a novel binary weight “significance” training scheme that fully capitalizes on the discrete nature of the optical spikes used by the SNN to process input information. The VCSEL-based photonic SNN was tested with a highly complex multivariate classification task (MADELON) before its performance was compared using a traditional least-squares training method and an alternative novel binary weighting scheme. Excellent classification accuracies of >94% were achieved by both training methods, exceeding the benchmark performance of the dataset in a fraction of the processing time. The newly reported training scheme also dramatically reduces the training set size requirements and the number of trained nodes (≤1% of the total network node count). This VCSEL-based photonic SNN, in combination with the reported “significance” weighting scheme, therefore grants ultrafast spike-based optical processing highly reduced training requirements and hardware complexity for potential application in future neuromorphic systems and artificial intelligence applications.","PeriodicalId":45291,"journal":{"name":"International Journal of Intelligent Computing and Cybernetics","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135685682","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hai Jin, Hao Qi, Jin Zhao, Xinyu Jiang, Yu Huang, Chuangyi Gui, Qinggang Wang, Xinyang Shen, Yi Zhang, Ao Hu, Dan Chen, Chao Liu, Haifeng Liu, Haiheng He, Xiangyu Ye, Runze Wang, Jingrui Yuan, Pengcheng Yao, Yu Zhang, Long Zheng, Xiaofei Liao
Graph analytics, which mainly includes graph processing, graph mining, and graph learning, has become increasingly important in several domains, including social network analysis, bioinformatics, and machine learning. However, graph analytics applications suffer from poor locality, limited bandwidth, and low parallelism owing to the irregular sparse structure, explosive growth, and dependencies of graph data. To address those challenges, several programming models, execution modes, and messaging strategies are proposed to improve the utilization of traditional hardware and performance. In recent years, novel computing and memory devices have emerged, e.g., HMCs, HBM, and ReRAM, providing massive bandwidth and parallelism resources, making it possible to address bottlenecks in graph applications. To facilitate understanding of the graph analytics domain, our study summarizes and categorizes current software systems implementation and domain-specific architectures. Finally, we discuss the future challenges of graph analytics.
{"title":"Software Systems Implementation and Domain-Specific Architectures towards Graph Analytics","authors":"Hai Jin, Hao Qi, Jin Zhao, Xinyu Jiang, Yu Huang, Chuangyi Gui, Qinggang Wang, Xinyang Shen, Yi Zhang, Ao Hu, Dan Chen, Chao Liu, Haifeng Liu, Haiheng He, Xiangyu Ye, Runze Wang, Jingrui Yuan, Pengcheng Yao, Yu Zhang, Long Zheng, Xiaofei Liao","doi":"10.34133/2022/9806758","DOIUrl":"https://doi.org/10.34133/2022/9806758","url":null,"abstract":"Graph analytics, which mainly includes graph processing, graph mining, and graph learning, has become increasingly important in several domains, including social network analysis, bioinformatics, and machine learning. However, graph analytics applications suffer from poor locality, limited bandwidth, and low parallelism owing to the irregular sparse structure, explosive growth, and dependencies of graph data. To address those challenges, several programming models, execution modes, and messaging strategies are proposed to improve the utilization of traditional hardware and performance. In recent years, novel computing and memory devices have emerged, e.g., HMCs, HBM, and ReRAM, providing massive bandwidth and parallelism resources, making it possible to address bottlenecks in graph applications. To facilitate understanding of the graph analytics domain, our study summarizes and categorizes current software systems implementation and domain-specific architectures. Finally, we discuss the future challenges of graph analytics.","PeriodicalId":45291,"journal":{"name":"International Journal of Intelligent Computing and Cybernetics","volume":"7 1","pages":""},"PeriodicalIF":4.3,"publicationDate":"2022-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87690495","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shixin Huang, Chao Chen, Gangya Zhu, Jinhan Xin, Z. Wang, Kai Hwang, Zhibin Yu
Stream data processing systems are becoming increasingly popular in the big data era. Systems such as Apache Flink typically provide a number (e.g., 30) of configuration parameters to flexibly specify the amount of resources (e.g., CPU cores and memory) allocated for tasks. These parameters significantly affect task performance. However, it is hard to manually tune them for optimal performance for an unknown program running on a given cluster. An automatic as well as fast resource configuration tuning approach is therefore desired. To this end, we propose to leverage Bayesian optimization to automatically tune the resource configurations for stream data processing systems. We first select a machine learning model—Random Forest—to construct accurate performance models for a stream data processing program. We subsequently take the Bayesian optimization (BO) algorithm, along with the performance models, to iteratively search the optimal configurations for a stream data processing program. Experimental results show that our approach improves the 99th-percentile tail latency by a factor of 2.62× on average and up to 5.26× overall. Furthermore, our approach improves throughput by a factor of 1.05× on average and up to 1.21× overall.
{"title":"Resource Configuration Tuning for Stream Data Processing Systems via Bayesian Optimization","authors":"Shixin Huang, Chao Chen, Gangya Zhu, Jinhan Xin, Z. Wang, Kai Hwang, Zhibin Yu","doi":"10.34133/2022/9820424","DOIUrl":"https://doi.org/10.34133/2022/9820424","url":null,"abstract":"Stream data processing systems are becoming increasingly popular in the big data era. Systems such as Apache Flink typically provide a number (e.g., 30) of configuration parameters to flexibly specify the amount of resources (e.g., CPU cores and memory) allocated for tasks. These parameters significantly affect task performance. However, it is hard to manually tune them for optimal performance for an unknown program running on a given cluster. An automatic as well as fast resource configuration tuning approach is therefore desired. To this end, we propose to leverage Bayesian optimization to automatically tune the resource configurations for stream data processing systems. We first select a machine learning model—Random Forest—to construct accurate performance models for a stream data processing program. We subsequently take the Bayesian optimization (BO) algorithm, along with the performance models, to iteratively search the optimal configurations for a stream data processing program. Experimental results show that our approach improves the 99th-percentile tail latency by a factor of 2.62× on average and up to 5.26× overall. Furthermore, our approach improves throughput by a factor of 1.05× on average and up to 1.21× overall.","PeriodicalId":45291,"journal":{"name":"International Journal of Intelligent Computing and Cybernetics","volume":"19 1","pages":""},"PeriodicalIF":4.3,"publicationDate":"2022-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86462802","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yanan Yang, Xiangyu Kong, Laiping Zhao, Yiming Li, Huanyu Zhang, Jie Li, Heng Qi, Keqiu Li
Colocating workloads are commonly used in datacenters to improve server utilization. However, the unpredictable application performance degradation caused by the contention for shared resources makes the problem difficult and limits the efficiency of this approach. This problem has sparked research in hardware and software techniques that focus on enhancing the datacenters’ isolation abilities. There is still lack of a comprehensive benchmark suite to evaluate such techniques. To address this problem, we present SDCBench, a new benchmark suite that is specifically designed for workload colocation and characterization in datacenters. SDCBench includes 16 applications that span a wide range of cloud scenarios, which are carefully selected from the existing benchmarks using the clustering analysis method. SDCBench implements a robust statistical methodology to support workload colocation and proposes a concept of latency entropy for measuring the isolation ability of cloud systems. It enables cloud tenants to understand the performance isolation ability in datacenters and choose their best-fitted cloud services. For cloud providers, it also helps them to improve the quality of service to increase their revenues. Experimental results show that SDCBench can simulate different workload colocation scenarios by generating pressures on multidimensional resources with simple configurations. We also use SDCBench to compare the latency entropies in public cloud platforms such as Huawei Cloud and AWS Cloud and a local prototype system FlameCluster-II; the evaluation results show FlameCluster-II has the best performance isolation ability over these three cloud systems, with 0.99 of experience availability and 0.29 of latency entropy.
{"title":"SDCBench: A Benchmark Suite for Workload Colocation and Evaluation in Datacenters","authors":"Yanan Yang, Xiangyu Kong, Laiping Zhao, Yiming Li, Huanyu Zhang, Jie Li, Heng Qi, Keqiu Li","doi":"10.34133/2022/9810691","DOIUrl":"https://doi.org/10.34133/2022/9810691","url":null,"abstract":"Colocating workloads are commonly used in datacenters to improve server utilization. However, the unpredictable application performance degradation caused by the contention for shared resources makes the problem difficult and limits the efficiency of this approach. This problem has sparked research in hardware and software techniques that focus on enhancing the datacenters’ isolation abilities. There is still lack of a comprehensive benchmark suite to evaluate such techniques. To address this problem, we present SDCBench, a new benchmark suite that is specifically designed for workload colocation and characterization in datacenters. SDCBench includes 16 applications that span a wide range of cloud scenarios, which are carefully selected from the existing benchmarks using the clustering analysis method. SDCBench implements a robust statistical methodology to support workload colocation and proposes a concept of latency entropy for measuring the isolation ability of cloud systems. It enables cloud tenants to understand the performance isolation ability in datacenters and choose their best-fitted cloud services. For cloud providers, it also helps them to improve the quality of service to increase their revenues. Experimental results show that SDCBench can simulate different workload colocation scenarios by generating pressures on multidimensional resources with simple configurations. We also use SDCBench to compare the latency entropies in public cloud platforms such as Huawei Cloud and AWS Cloud and a local prototype system FlameCluster-II; the evaluation results show FlameCluster-II has the best performance isolation ability over these three cloud systems, with 0.99 of experience availability and 0.29 of latency entropy.","PeriodicalId":45291,"journal":{"name":"International Journal of Intelligent Computing and Cybernetics","volume":"6 1","pages":""},"PeriodicalIF":4.3,"publicationDate":"2022-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85238626","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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：