Pub Date : 2016-06-27DOI: 10.1109/ISCC.2016.7543840
Gulnur Selda Uyanik, S. Oktug
Cognitive radio networks (CRNs) are composed of unprivileged users as called secondary users (SUs) that utilize the spectrum opportunities produced by the absence of co-located privileged primary users (PUs). Providing a certain level of quality of service (QoS) to these SUs is a very challenging problem as these users are constrained strictly by PU arrivals. Many existing works tackled this problem by centralized and complex optimization techniques and complex mac layer protocols. As opposed to the existing QoS management schemes, in this paper, we propose a fast, distributed, PU temporal activity estimation aided spectrum assignment scheme for the SUs with traffic demands at different QoS levels. We designed our heuristic method by leveraging low prioritized (LP) packets over high prioritized (HP) packets on channels in heavy use in terms of temporal PU activity and used the opposite manner for the channels being in sparse use. We modeled our proposed scheme along with the network model we employed in MATLAB® and evaluated its performance with respect to a native scheme that utilities every spectrum opportunity for the HP traffic. Our simulations validated the effectiveness of the proposed channel assignment scheme in terms of Figure of Merit metric we defined reflecting the weighted sum of the throughput ratio values for packets of different QoS levels that successfully reached at the destination SU.
{"title":"Primary user activity classification aided channel assignment in cognitive radio networks","authors":"Gulnur Selda Uyanik, S. Oktug","doi":"10.1109/ISCC.2016.7543840","DOIUrl":"https://doi.org/10.1109/ISCC.2016.7543840","url":null,"abstract":"Cognitive radio networks (CRNs) are composed of unprivileged users as called secondary users (SUs) that utilize the spectrum opportunities produced by the absence of co-located privileged primary users (PUs). Providing a certain level of quality of service (QoS) to these SUs is a very challenging problem as these users are constrained strictly by PU arrivals. Many existing works tackled this problem by centralized and complex optimization techniques and complex mac layer protocols. As opposed to the existing QoS management schemes, in this paper, we propose a fast, distributed, PU temporal activity estimation aided spectrum assignment scheme for the SUs with traffic demands at different QoS levels. We designed our heuristic method by leveraging low prioritized (LP) packets over high prioritized (HP) packets on channels in heavy use in terms of temporal PU activity and used the opposite manner for the channels being in sparse use. We modeled our proposed scheme along with the network model we employed in MATLAB® and evaluated its performance with respect to a native scheme that utilities every spectrum opportunity for the HP traffic. Our simulations validated the effectiveness of the proposed channel assignment scheme in terms of Figure of Merit metric we defined reflecting the weighted sum of the throughput ratio values for packets of different QoS levels that successfully reached at the destination SU.","PeriodicalId":148096,"journal":{"name":"2016 IEEE Symposium on Computers and Communication (ISCC)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132281393","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 : 2016-06-27DOI: 10.1109/ISCC.2016.7543846
Shuzhi Su, H. Ge, Yunhao Yuan
Locality-based feature learning has drawn more and more attentions recently. However, most of locality-based feature learning methods only consider a kind of local neighbor information, and such the locality-based methods are difficult to well reveal intrinsic geometrical structure of raw high-dimensional data. In this paper, we propose a novel multi-locality correlation feature learning algorithm for multi-view data, called multi-locality discrimination canonical correlation analysis (MLDCCA), which can learn nonlinear correlation features with strong discriminative power. Different from the locality-based methods, our algorithm not only employs multiple local patches of each raw data to well capture the intrinsic geometrical structure information, but also fully considers intraclass scatter information for further enhancing the class separability of the learned correlation features. Extensive experimental results on several real-word image datasets have demonstrated the effectiveness of our algorithm.
{"title":"Multi-locality correlation feature learning for image recognition","authors":"Shuzhi Su, H. Ge, Yunhao Yuan","doi":"10.1109/ISCC.2016.7543846","DOIUrl":"https://doi.org/10.1109/ISCC.2016.7543846","url":null,"abstract":"Locality-based feature learning has drawn more and more attentions recently. However, most of locality-based feature learning methods only consider a kind of local neighbor information, and such the locality-based methods are difficult to well reveal intrinsic geometrical structure of raw high-dimensional data. In this paper, we propose a novel multi-locality correlation feature learning algorithm for multi-view data, called multi-locality discrimination canonical correlation analysis (MLDCCA), which can learn nonlinear correlation features with strong discriminative power. Different from the locality-based methods, our algorithm not only employs multiple local patches of each raw data to well capture the intrinsic geometrical structure information, but also fully considers intraclass scatter information for further enhancing the class separability of the learned correlation features. Extensive experimental results on several real-word image datasets have demonstrated the effectiveness of our algorithm.","PeriodicalId":148096,"journal":{"name":"2016 IEEE Symposium on Computers and Communication (ISCC)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133880927","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 : 2016-06-27DOI: 10.1109/ISCC.2016.7543891
Jinghong Wu, Hong Ni, Xuewen Zeng, Liping Ding, Xiaozhou Ye
OpenFlow, which facilitates decoupling between the forwarding and control plane, is developing rapidly and has already been widely studied in various fields. OpenFlow switches perform certain actions under the guidance of flow tables, which are configured by the controller. Currently, most of the work based on OpenFlow switch is concerned with switching performance such as forwarding rate, forwarding latency and flow table efficiency. In contrast, little work has been done on storage for OpenFlow switches. However, directly imposing storage functionality on OpenFlow switches with large capacity storage devices will lead to a series of problems. The most protruding and apparent one is the degradation of forwarding rate, which is one of the most important figure of merits for OpenFlow switch. This paper analyses the typical problems in this context and proposes a novel storage approach based on Protocol Oblivious Forwarding (POF), which is an enhancement to OpenFlow-based SDN forwarding architecture. The preliminary experimental results on Linux-based POF soft switch validate the effectiveness and efficiency of our approach.
{"title":"A storage approach for OpenFlow switch based on Protocol Oblivious Forwarding","authors":"Jinghong Wu, Hong Ni, Xuewen Zeng, Liping Ding, Xiaozhou Ye","doi":"10.1109/ISCC.2016.7543891","DOIUrl":"https://doi.org/10.1109/ISCC.2016.7543891","url":null,"abstract":"OpenFlow, which facilitates decoupling between the forwarding and control plane, is developing rapidly and has already been widely studied in various fields. OpenFlow switches perform certain actions under the guidance of flow tables, which are configured by the controller. Currently, most of the work based on OpenFlow switch is concerned with switching performance such as forwarding rate, forwarding latency and flow table efficiency. In contrast, little work has been done on storage for OpenFlow switches. However, directly imposing storage functionality on OpenFlow switches with large capacity storage devices will lead to a series of problems. The most protruding and apparent one is the degradation of forwarding rate, which is one of the most important figure of merits for OpenFlow switch. This paper analyses the typical problems in this context and proposes a novel storage approach based on Protocol Oblivious Forwarding (POF), which is an enhancement to OpenFlow-based SDN forwarding architecture. The preliminary experimental results on Linux-based POF soft switch validate the effectiveness and efficiency of our approach.","PeriodicalId":148096,"journal":{"name":"2016 IEEE Symposium on Computers and Communication (ISCC)","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114364267","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 : 2016-06-27DOI: 10.1109/ISCC.2016.7543792
J. Ceron, C. Margi, L. Granville
Mechanisms to detect and analyze malicious software are essential to improve security systems. Current security mechanisms have limited success in detecting sophisticated malicious software. More than to evade analysis system, many malwares require specific conditions to activate their actions in the target system. The flexibility of Software-Defined Networking (SDN) provides an opportunity to develop a malware analysis architecture integrating different systems and networks profile configuration. In this paper we design an architecture specialized in malware analysis using SDN to dynamically reconfigure the network environment based on malware actions. As result, we demonstrate that our solution can trigger more malware's events than traditional solutions that do not consider sandbox surround environment as an important component in malware analysis.
{"title":"MARS: An SDN-based malware analysis solution","authors":"J. Ceron, C. Margi, L. Granville","doi":"10.1109/ISCC.2016.7543792","DOIUrl":"https://doi.org/10.1109/ISCC.2016.7543792","url":null,"abstract":"Mechanisms to detect and analyze malicious software are essential to improve security systems. Current security mechanisms have limited success in detecting sophisticated malicious software. More than to evade analysis system, many malwares require specific conditions to activate their actions in the target system. The flexibility of Software-Defined Networking (SDN) provides an opportunity to develop a malware analysis architecture integrating different systems and networks profile configuration. In this paper we design an architecture specialized in malware analysis using SDN to dynamically reconfigure the network environment based on malware actions. As result, we demonstrate that our solution can trigger more malware's events than traditional solutions that do not consider sandbox surround environment as an important component in malware analysis.","PeriodicalId":148096,"journal":{"name":"2016 IEEE Symposium on Computers and Communication (ISCC)","volume":"2016 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114455601","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 : 2016-06-27DOI: 10.1109/ISCC.2016.7543867
Carlos Roberto de Rolt, R. Montanari, Marcelo Luiz Brocardo, L. Foschini, J. S. Dias
Recent advancements in wireless technologies and the widespread availability of smartphones equipped with several physical and virtual sensors are leading to the emergence of novel mobile healthcare scenarios where patients with critical physical/behavioral conditions can be provided with anywhere and anytime care assistance even while on the move. Crowdsensing, through the massive use of smartphone sensors further enhances the potential of supporting participatory management of emergency scenarios. This paper presents a macro process modelling of crowdsourced-based participatory emergency scenarios and, accordingly, proposes a crowdsensing-based middleware called COLLEGA that provides comprehensive management functionalities for supporting prompt assistance in case of a medical emergency. In particular, through participatory and opportunistic sensing, the COLLEGA framework allows dynamic formation of ad-hoc assistance groups formed by passing-by users capable of assisting mobile patients in need of help while waiting for professional caregivers and provides support for understanding the emergency situation and effectively planning and executing assistance actions.
{"title":"COLLEGA middleware for the management of participatory Mobile Health Communities","authors":"Carlos Roberto de Rolt, R. Montanari, Marcelo Luiz Brocardo, L. Foschini, J. S. Dias","doi":"10.1109/ISCC.2016.7543867","DOIUrl":"https://doi.org/10.1109/ISCC.2016.7543867","url":null,"abstract":"Recent advancements in wireless technologies and the widespread availability of smartphones equipped with several physical and virtual sensors are leading to the emergence of novel mobile healthcare scenarios where patients with critical physical/behavioral conditions can be provided with anywhere and anytime care assistance even while on the move. Crowdsensing, through the massive use of smartphone sensors further enhances the potential of supporting participatory management of emergency scenarios. This paper presents a macro process modelling of crowdsourced-based participatory emergency scenarios and, accordingly, proposes a crowdsensing-based middleware called COLLEGA that provides comprehensive management functionalities for supporting prompt assistance in case of a medical emergency. In particular, through participatory and opportunistic sensing, the COLLEGA framework allows dynamic formation of ad-hoc assistance groups formed by passing-by users capable of assisting mobile patients in need of help while waiting for professional caregivers and provides support for understanding the emergency situation and effectively planning and executing assistance actions.","PeriodicalId":148096,"journal":{"name":"2016 IEEE Symposium on Computers and Communication (ISCC)","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115905897","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 : 2016-06-27DOI: 10.1109/ISCC.2016.7543856
Yuncheng Wu, Juru Zeng, Hong Chen, Yao Wu, Wenjuan Liang, Hui Peng, Cuiping Li
Recently, listwise ranking-oriented collaborative filtering (CF) algorithms have gained great success in recommender systems. However, the ranked preference list may compromise the privacy of individuals. A notable paradigm for offering strong privacy guarantee is differential privacy. In this paper, we propose DPListCF, a differentially private algorithm based on ListCF (a state-of-art listwise CF algorithm). The main idea of DPListCF is to make both of the similarity calculation phase and rank prediction phase of ListCF satisfy differential privacy, by using input perturbation method and output perturbation method in the two phases respectively. Extensive experiments using two real datasets evaluate the performance of DPListCF, and demonstrate that the proposed algorithm outperforms state-of-art approaches.
{"title":"DPListCF: A differentially private approach for listwise collaborative filtering","authors":"Yuncheng Wu, Juru Zeng, Hong Chen, Yao Wu, Wenjuan Liang, Hui Peng, Cuiping Li","doi":"10.1109/ISCC.2016.7543856","DOIUrl":"https://doi.org/10.1109/ISCC.2016.7543856","url":null,"abstract":"Recently, listwise ranking-oriented collaborative filtering (CF) algorithms have gained great success in recommender systems. However, the ranked preference list may compromise the privacy of individuals. A notable paradigm for offering strong privacy guarantee is differential privacy. In this paper, we propose DPListCF, a differentially private algorithm based on ListCF (a state-of-art listwise CF algorithm). The main idea of DPListCF is to make both of the similarity calculation phase and rank prediction phase of ListCF satisfy differential privacy, by using input perturbation method and output perturbation method in the two phases respectively. Extensive experiments using two real datasets evaluate the performance of DPListCF, and demonstrate that the proposed algorithm outperforms state-of-art approaches.","PeriodicalId":148096,"journal":{"name":"2016 IEEE Symposium on Computers and Communication (ISCC)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115164120","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 : 2016-06-27DOI: 10.1109/ISCC.2016.7543709
S. Cuomo, P. D. Michele, A. Galletti, L. Marcellino
We focus on the Overcomplete Local Principal Component Analysis (OLPCA) method, which is widely adopted as denoising filter. We propose a programming approach resorting to Graphic Processor Units (GPUs), in order to massively parallelize some heavy computational tasks of the method. In our approach, we design and implement a parallel version of the OLPCA, by using a suitable mapping of the tasks on a GPU architecture with the aim to investigate the performance and the denoising features of the algorithm. The experimental results show improvements in terms of GFlops and memory throughput.
{"title":"A GPU parallel implementation of the Local Principal Component Analysis overcomplete method for DW image denoising","authors":"S. Cuomo, P. D. Michele, A. Galletti, L. Marcellino","doi":"10.1109/ISCC.2016.7543709","DOIUrl":"https://doi.org/10.1109/ISCC.2016.7543709","url":null,"abstract":"We focus on the Overcomplete Local Principal Component Analysis (OLPCA) method, which is widely adopted as denoising filter. We propose a programming approach resorting to Graphic Processor Units (GPUs), in order to massively parallelize some heavy computational tasks of the method. In our approach, we design and implement a parallel version of the OLPCA, by using a suitable mapping of the tasks on a GPU architecture with the aim to investigate the performance and the denoising features of the algorithm. The experimental results show improvements in terms of GFlops and memory throughput.","PeriodicalId":148096,"journal":{"name":"2016 IEEE Symposium on Computers and Communication (ISCC)","volume":"315 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124464880","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 : 2016-06-27DOI: 10.1109/ISCC.2016.7543768
Flaviano Di Rienzo, M. Girolami, S. Chessa, F. Paparella, A. Caruso
Underwater communications through acoustic modems rise several networking challenges for the Underwater Acoustic Sensor Networks (UASN). In particular, opportunistic routing is a novel but promising technique that can remarkably increase the reliability of the UASN, but its use in this context requires studies on the nature of mobility in UASN. Our goal is to study a real-world mobility dataset obtained from the Argo project. In particular, we observe the mobility of 51 free-drifting floats deployed on the Mediterranean Sea for approximately one year and we analyze some important properties of the underwater network we built. Specifically, we analyze the contact-time, inter-contact time as well density and network degree while varying the connectivity degree of the whole dataset. We then consider three known routing algorithms, namely Epidemic, PROPHET and Direct Delivery, with the goal of measuring their performance in real conditions for USAN. We finally discuss the opportunities arising from the adoption of opportunistic routing in UASN showing that, even in a very sparse and strongly disconnected network, it is still possible to build a limited but working networking framework.
{"title":"Signals from the depths: Properties of percolation strategies with the Argo dataset","authors":"Flaviano Di Rienzo, M. Girolami, S. Chessa, F. Paparella, A. Caruso","doi":"10.1109/ISCC.2016.7543768","DOIUrl":"https://doi.org/10.1109/ISCC.2016.7543768","url":null,"abstract":"Underwater communications through acoustic modems rise several networking challenges for the Underwater Acoustic Sensor Networks (UASN). In particular, opportunistic routing is a novel but promising technique that can remarkably increase the reliability of the UASN, but its use in this context requires studies on the nature of mobility in UASN. Our goal is to study a real-world mobility dataset obtained from the Argo project. In particular, we observe the mobility of 51 free-drifting floats deployed on the Mediterranean Sea for approximately one year and we analyze some important properties of the underwater network we built. Specifically, we analyze the contact-time, inter-contact time as well density and network degree while varying the connectivity degree of the whole dataset. We then consider three known routing algorithms, namely Epidemic, PROPHET and Direct Delivery, with the goal of measuring their performance in real conditions for USAN. We finally discuss the opportunities arising from the adoption of opportunistic routing in UASN showing that, even in a very sparse and strongly disconnected network, it is still possible to build a limited but working networking framework.","PeriodicalId":148096,"journal":{"name":"2016 IEEE Symposium on Computers and Communication (ISCC)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128570841","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 : 2016-06-27DOI: 10.1109/ISCC.2016.7543752
F. Murgia, I. Tagliente, I. Zoppis, G. Mauri, F. Sicurello, F. Bella, Vanessa Mercuri, E. Santoro, G. Castelnuovo, S. Bella
Since 2001, in the Cystic Fibrosis Center of the Pediatric Hospital Bambino Gesù in Rome, we use telemedicine for monitoring of our patients. While in our first published works reporting this experience, we showed statistically significant reduction in hospital admissions and a tendency over time towards a better stability of the respiratory function for telehomecare (THC) patients, here we focus on the trend of the Forced Expiratory Volume in the first second (FEV1). In particular, we investigate the evolution of the clinical trend of the FEV1 index, by monitoring the activities of home patients from 2011 to 2014. THC is applied in addition to the standard therapeutic protocol by following 16 Cystic Fibrosis (CF) patients with specialized doctors. Our results show that THC patients improve their FEV1 values with a trend which can be considered significantly better than the one reported by the control group.
{"title":"Trend of FEV1 in Cystic Fibrosis patients: A telehomecare experience","authors":"F. Murgia, I. Tagliente, I. Zoppis, G. Mauri, F. Sicurello, F. Bella, Vanessa Mercuri, E. Santoro, G. Castelnuovo, S. Bella","doi":"10.1109/ISCC.2016.7543752","DOIUrl":"https://doi.org/10.1109/ISCC.2016.7543752","url":null,"abstract":"Since 2001, in the Cystic Fibrosis Center of the Pediatric Hospital Bambino Gesù in Rome, we use telemedicine for monitoring of our patients. While in our first published works reporting this experience, we showed statistically significant reduction in hospital admissions and a tendency over time towards a better stability of the respiratory function for telehomecare (THC) patients, here we focus on the trend of the Forced Expiratory Volume in the first second (FEV1). In particular, we investigate the evolution of the clinical trend of the FEV1 index, by monitoring the activities of home patients from 2011 to 2014. THC is applied in addition to the standard therapeutic protocol by following 16 Cystic Fibrosis (CF) patients with specialized doctors. Our results show that THC patients improve their FEV1 values with a trend which can be considered significantly better than the one reported by the control group.","PeriodicalId":148096,"journal":{"name":"2016 IEEE Symposium on Computers and Communication (ISCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130393996","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 : 2016-06-27DOI: 10.1109/ISCC.2016.7543829
Rashid Morady, D. Dal
Multiprocessor task scheduling is one of the hardest combinatorial optimization problems in parallel and distributed systems. It is known as NP-hard and therefore, scanning the whole search space using an exact algorithm to find the optimal solution is not practical. Instead, metaheuristics are mostly employed to find a near-optimal solution in a reasonable amount of time. In this paper, a multi-population based parallel genetic algorithm is presented for the optimization of multiprocessor task scheduling in the presence of communication costs. To the best of our knowledge, this parallel genetic algorithm approach is applied to the problem at hand for the first time using a benchmark set that includes well-known task graphs from different sources. Our experiments conducted on several task graphs with different sizes from the benchmark set showed the superiority of the approach over a conventional genetic algorithm and the related works in the literature in terms of two different performance metrics. Our parallel implementation not only decreased the execution time but also increased the quality of the scheduling solutions considerably.
{"title":"A multi-population based parallel genetic algorithm for multiprocessor task scheduling with Communication Costs","authors":"Rashid Morady, D. Dal","doi":"10.1109/ISCC.2016.7543829","DOIUrl":"https://doi.org/10.1109/ISCC.2016.7543829","url":null,"abstract":"Multiprocessor task scheduling is one of the hardest combinatorial optimization problems in parallel and distributed systems. It is known as NP-hard and therefore, scanning the whole search space using an exact algorithm to find the optimal solution is not practical. Instead, metaheuristics are mostly employed to find a near-optimal solution in a reasonable amount of time. In this paper, a multi-population based parallel genetic algorithm is presented for the optimization of multiprocessor task scheduling in the presence of communication costs. To the best of our knowledge, this parallel genetic algorithm approach is applied to the problem at hand for the first time using a benchmark set that includes well-known task graphs from different sources. Our experiments conducted on several task graphs with different sizes from the benchmark set showed the superiority of the approach over a conventional genetic algorithm and the related works in the literature in terms of two different performance metrics. Our parallel implementation not only decreased the execution time but also increased the quality of the scheduling solutions considerably.","PeriodicalId":148096,"journal":{"name":"2016 IEEE Symposium on Computers and Communication (ISCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128675581","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：