Morikawa Hiroaki, H. Ebara, Onishi Katsumi, Nakano Hideo
Nowadays, computing resources have taken tens percent of utilization at busy time for increasing computing power. However, the virtual machine (VM) technology is effective to the use of computing resources. Besides, the vulnerable services in VM prevent the affect of the other VM attacks. For example, Amazon has applied the VM method to run in real machine (Host OS) independently but attackers can retrieve data by CPU and memory dump software. Therefore this study suggests a method to improve the access virtual Trusted Platform Module (vTPM) implemented in Xen software to keep the server system safely. Furthermore, we implement and evaluate this method.
{"title":"Improvement for vTPM Access Control on Xen","authors":"Morikawa Hiroaki, H. Ebara, Onishi Katsumi, Nakano Hideo","doi":"10.1109/ICPPW.2010.44","DOIUrl":"https://doi.org/10.1109/ICPPW.2010.44","url":null,"abstract":"Nowadays, computing resources have taken tens percent of utilization at busy time for increasing computing power. However, the virtual machine (VM) technology is effective to the use of computing resources. Besides, the vulnerable services in VM prevent the affect of the other VM attacks. For example, Amazon has applied the VM method to run in real machine (Host OS) independently but attackers can retrieve data by CPU and memory dump software. Therefore this study suggests a method to improve the access virtual Trusted Platform Module (vTPM) implemented in Xen software to keep the server system safely. Furthermore, we implement and evaluate this method.","PeriodicalId":415472,"journal":{"name":"2010 39th International Conference on Parallel Processing Workshops","volume":"15 10","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131574168","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}
Green computing is a hot topic that received a great amount of interest in the past few years. This paper explores the benefits of Dynamic Voltage/Frequency Scaling (DV/FS) and server number controlling Vary-On Vary-Off (VOVF) for power management in a server cluster. Previous work mostly addressed the DV/FS and VOVF separately. In this paper, an optimization problem is formulated to achieve energy savings while ensure performance. DV/FS and VOVF mechanisms are combined to obtain an optimal power management tactics. The presented optimization model provides controllable and predictable quantitative control of power consumption with theoretically guaranteed service performance. We further study overhead of the proposed scheme and provide a Double Control Periods (DCP) model to compensate the transition overhead. The power optimization model and DCP model are evaluated via extensive simulations, they are also justified by the real workload data trace. The results prove the effectiveness and efficiency of the proposed models.
{"title":"Optimal Server Provisioning and Frequency Adjustment in Server Clusters","authors":"Xinying Zheng, Yu Cai","doi":"10.1109/ICPPW.2010.74","DOIUrl":"https://doi.org/10.1109/ICPPW.2010.74","url":null,"abstract":"Green computing is a hot topic that received a great amount of interest in the past few years. This paper explores the benefits of Dynamic Voltage/Frequency Scaling (DV/FS) and server number controlling Vary-On Vary-Off (VOVF) for power management in a server cluster. Previous work mostly addressed the DV/FS and VOVF separately. In this paper, an optimization problem is formulated to achieve energy savings while ensure performance. DV/FS and VOVF mechanisms are combined to obtain an optimal power management tactics. The presented optimization model provides controllable and predictable quantitative control of power consumption with theoretically guaranteed service performance. We further study overhead of the proposed scheme and provide a Double Control Periods (DCP) model to compensate the transition overhead. The power optimization model and DCP model are evaluated via extensive simulations, they are also justified by the real workload data trace. The results prove the effectiveness and efficiency of the proposed models.","PeriodicalId":415472,"journal":{"name":"2010 39th International Conference on Parallel Processing Workshops","volume":"520 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116264560","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}
Modern data centers must provide performance assurance for complex system software such as multi-tier web applications. In addition, the power consumption of data centers needs to be minimized to reduce operating costs and avoid system overheating. Various power-efficient performance management strategies have been proposed based on dynamic voltage and frequency scaling (DVFS). Virtualization technologies have also made it possible to consolidate multiple virtual machines (VMs) onto a smaller number of active physical servers for even greater power savings, but at the cost of a higher overhead. This paper proposes a performance-controlled power optimization solution for virtualized data centers with multi-tier applications. While existing work relies on either DVFS or server consolidation in a separate manner, our solution utilizes both strategies for maximized power savings by integrating feedback control with optimization strategies. At the application level, a multi-input-multi-output controller is designed to achieve the desired performance for applications spanning multiple VMs, on a short time scale, by reallocating the CPU resources and DVFS. At the data center level, a power optimizer is proposed to incrementally consolidate VMs onto the most power-efficient servers on a longer time scale. Empirical results on a hardware testbed demonstrate that our solution can effectively achieve performance-assured power savings. Extensive simulation results, based on a trace file of 5,415 real servers, demonstrate the efficacy of our solution in large-scale data centers.
{"title":"Power Optimization with Performance Assurance for Multi-tier Applications in Virtualized Data Centers","authors":"Yefu Wang, Xiaorui Wang","doi":"10.1109/ICPPW.2010.75","DOIUrl":"https://doi.org/10.1109/ICPPW.2010.75","url":null,"abstract":"Modern data centers must provide performance assurance for complex system software such as multi-tier web applications. In addition, the power consumption of data centers needs to be minimized to reduce operating costs and avoid system overheating. Various power-efficient performance management strategies have been proposed based on dynamic voltage and frequency scaling (DVFS). Virtualization technologies have also made it possible to consolidate multiple virtual machines (VMs) onto a smaller number of active physical servers for even greater power savings, but at the cost of a higher overhead. This paper proposes a performance-controlled power optimization solution for virtualized data centers with multi-tier applications. While existing work relies on either DVFS or server consolidation in a separate manner, our solution utilizes both strategies for maximized power savings by integrating feedback control with optimization strategies. At the application level, a multi-input-multi-output controller is designed to achieve the desired performance for applications spanning multiple VMs, on a short time scale, by reallocating the CPU resources and DVFS. At the data center level, a power optimizer is proposed to incrementally consolidate VMs onto the most power-efficient servers on a longer time scale. Empirical results on a hardware testbed demonstrate that our solution can effectively achieve performance-assured power savings. Extensive simulation results, based on a trace file of 5,415 real servers, demonstrate the efficacy of our solution in large-scale data centers.","PeriodicalId":415472,"journal":{"name":"2010 39th International Conference on Parallel Processing Workshops","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122101973","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}
Data storage technologies have been recognized as one of the major dimensions of information management along with the network infrastructure and applications. The prosperity of cloud computing requires the migration from server-attached storage to network-based distributed storage. Along with variant advantages, distributed storage also poses new challenges in creating a secure and reliable data storage and access facility. The data security in cloud is one of the challenges to be addressed before the novel pay-as-you-go business model can be accepted and applied widely. Concerns are raised from both insecure/unreliable service providers and potential malicious users. In this article, we analyze the integrity vulnerability existing in the current cloud storage platforms and show the problem of repudiation. A novel non-repudiation (NR) protocol specifically designed in the context of cloud computing environment is proposed. We have also discussed the robustness of the NR protocol against typical attacks in the network environments.
{"title":"Analysis of Integrity Vulnerabilities and a Non-repudiation Protocol for Cloud Data Storage Platforms","authors":"Jun Feng, Yu Chen, Wei-Shinn Ku, Pu Liu","doi":"10.1109/ICPPW.2010.42","DOIUrl":"https://doi.org/10.1109/ICPPW.2010.42","url":null,"abstract":"Data storage technologies have been recognized as one of the major dimensions of information management along with the network infrastructure and applications. The prosperity of cloud computing requires the migration from server-attached storage to network-based distributed storage. Along with variant advantages, distributed storage also poses new challenges in creating a secure and reliable data storage and access facility. The data security in cloud is one of the challenges to be addressed before the novel pay-as-you-go business model can be accepted and applied widely. Concerns are raised from both insecure/unreliable service providers and potential malicious users. In this article, we analyze the integrity vulnerability existing in the current cloud storage platforms and show the problem of repudiation. A novel non-repudiation (NR) protocol specifically designed in the context of cloud computing environment is proposed. We have also discussed the robustness of the NR protocol against typical attacks in the network environments.","PeriodicalId":415472,"journal":{"name":"2010 39th International Conference on Parallel Processing Workshops","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129839327","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}
Traditional operating systems and applications use logs extensively to monitor system activity and perform intrusion detection. Consequently, logs have also become prime targets for intruders. When a malware or intruder obtains root privileges in a system, one of its first actions is to hide its footprint by deleting or modifying system logs, especially the log entry recording the intrusion activity (such as unauthorized root login). A key weakness of most current logging mechanisms is that logs are stored on a storage device over which the system being logged has complete control, including the ability to delete/modify the logs arbitrarily. Once the root privileges of such a system are compromised, so are the logs. Virtualization offers a unique opportunity to eliminate this point of weakness. In this paper, we propose a new virtual storage abstraction for virtual machines (VMs) called Virtual Append-only Storage (VAS) that secures and preserves all system and/or application logs in a VM and can prevent an intruder from deleting/modifying past logs even after the root privileges of a VM are compromised. Our VAS-based logging complements existing intrusion detection techniques which mainly monitor the in-memory execution state and data, but do not protect the storage device on which logs are stored. Since logs can become voluminous over time, VAS also provides administrators the ability to secure either system-wide or application-specific logs, rather than blindly logging all system activity.
{"title":"A Case for Secure Virtual Append-Only Storage for Virtual Machines","authors":"Zhao Lin, Kartik Gopalan, Ping Yang","doi":"10.1109/ICPPW.2010.15","DOIUrl":"https://doi.org/10.1109/ICPPW.2010.15","url":null,"abstract":"Traditional operating systems and applications use logs extensively to monitor system activity and perform intrusion detection. Consequently, logs have also become prime targets for intruders. When a malware or intruder obtains root privileges in a system, one of its first actions is to hide its footprint by deleting or modifying system logs, especially the log entry recording the intrusion activity (such as unauthorized root login). A key weakness of most current logging mechanisms is that logs are stored on a storage device over which the system being logged has complete control, including the ability to delete/modify the logs arbitrarily. Once the root privileges of such a system are compromised, so are the logs. Virtualization offers a unique opportunity to eliminate this point of weakness. In this paper, we propose a new virtual storage abstraction for virtual machines (VMs) called Virtual Append-only Storage (VAS) that secures and preserves all system and/or application logs in a VM and can prevent an intruder from deleting/modifying past logs even after the root privileges of a VM are compromised. Our VAS-based logging complements existing intrusion detection techniques which mainly monitor the in-memory execution state and data, but do not protect the storage device on which logs are stored. Since logs can become voluminous over time, VAS also provides administrators the ability to secure either system-wide or application-specific logs, rather than blindly logging all system activity.","PeriodicalId":415472,"journal":{"name":"2010 39th International Conference on Parallel Processing Workshops","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115196770","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}
Modern HPC systems are constructed by placing more and more cores in a single machine. To utilize this kind of machines efficiently, many parallel processes have to be used. The performance analysis of massively parallel program runs becomes more and more complicated as the number of events that are generated while the program is traced grows linearly with the number of processes. In order to utilize large HPC systems efficiently, parallel applications have to execute I/O requests in parallel. Analyzing these I/O requests and optimizing this part of a parallel program requires a deep knowledge of all issued requests and the dependencies between these requests. Traditional tracing facilities record all necessary information, including all synchronization events. We present a novel approach to reduce the amount of information needed for an I/O analysis in program traces significantly. This reduction enables a further analysis of the reduced data set in other tools that for example detect request patterns. Our approach is based on a specialized graph that is constructed from an event trace. This paper describes a systematic methodology to reduce the initial graph by merging adjacent vertices. As an extension we also describe how this merging step can be combined with the graph construction which significantly reduces the runtime of the algorithm in practice. An example that demonstrates the practical application of the methodology to real world use cases concludes the paper. After applying the reduction operation to application traces in the example the amount of synchronization events remaining is in the order of the number of I/O events.
{"title":"Efficient Pattern Based I/O Analysis of Parallel Programs","authors":"Michael Kluge, A. Knüpfer, W. Nagel","doi":"10.1109/ICPPW.2010.31","DOIUrl":"https://doi.org/10.1109/ICPPW.2010.31","url":null,"abstract":"Modern HPC systems are constructed by placing more and more cores in a single machine. To utilize this kind of machines efficiently, many parallel processes have to be used. The performance analysis of massively parallel program runs becomes more and more complicated as the number of events that are generated while the program is traced grows linearly with the number of processes. In order to utilize large HPC systems efficiently, parallel applications have to execute I/O requests in parallel. Analyzing these I/O requests and optimizing this part of a parallel program requires a deep knowledge of all issued requests and the dependencies between these requests. Traditional tracing facilities record all necessary information, including all synchronization events. We present a novel approach to reduce the amount of information needed for an I/O analysis in program traces significantly. This reduction enables a further analysis of the reduced data set in other tools that for example detect request patterns. Our approach is based on a specialized graph that is constructed from an event trace. This paper describes a systematic methodology to reduce the initial graph by merging adjacent vertices. As an extension we also describe how this merging step can be combined with the graph construction which significantly reduces the runtime of the algorithm in practice. An example that demonstrates the practical application of the methodology to real world use cases concludes the paper. After applying the reduction operation to application traces in the example the amount of synchronization events remaining is in the order of the number of I/O events.","PeriodicalId":415472,"journal":{"name":"2010 39th International Conference on Parallel Processing Workshops","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132638641","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}
C. Grimme, Joachim Lepping, Jonathan Moreno Picon, A. Papaspyrou
In this paper, we propose a new algorithm for job interchange in Computational Grids that consist of autonomous and equitable HPC sites, called Shaking-G. Originally developed for balancing the sharing of video files in P2P networks, we conceptually transfer and adapt the algorithm to the domain of job scheduling in Grids, building an integrated, load-adaptive two-tier job exchange strategy. We evaluate the performance of Shaking-G with real workload data in different experimental scenarios and show that it outperforms traditional non-Grid aware algorithms in setups without job interchange, fostering the benefits of collaboration between large HPC centers.
{"title":"Applying P2P Strategies to Scheduling in Decentralized Grid Computing Infrastructures","authors":"C. Grimme, Joachim Lepping, Jonathan Moreno Picon, A. Papaspyrou","doi":"10.1109/ICPPW.2010.47","DOIUrl":"https://doi.org/10.1109/ICPPW.2010.47","url":null,"abstract":"In this paper, we propose a new algorithm for job interchange in Computational Grids that consist of autonomous and equitable HPC sites, called Shaking-G. Originally developed for balancing the sharing of video files in P2P networks, we conceptually transfer and adapt the algorithm to the domain of job scheduling in Grids, building an integrated, load-adaptive two-tier job exchange strategy. We evaluate the performance of Shaking-G with real workload data in different experimental scenarios and show that it outperforms traditional non-Grid aware algorithms in setups without job interchange, fostering the benefits of collaboration between large HPC centers.","PeriodicalId":415472,"journal":{"name":"2010 39th International Conference on Parallel Processing Workshops","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131158343","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}
The innovative exploitation of network coding technology could bring great design advantages to MANETs. First, wireless links are inherently lossy due to channel fading or interference. Second, the delivery condition of the route from the source to each receiver in a multicast transmission might be significantly distinct. Trying to satisfy the reliability requirement for the poorly-connected receivers may affect the performance of the rest of the receivers. In this paper, we present a tree-based multicast protocol, which exploits the characteristics of network coding to provide efficient and reliable multicast. A metric, named Receiving Probability (RP), is proposed to represent the probability for a node in a multicast tree to successfully receive a packet sent by the multicast source. To mitigate the distinctions among the receivers, the proposed multicast protocol aims to construct multiple trees such that the RP of each receiver satisfies a predefined threshold. Besides, since seldom redundant packets are generated with random network coding, by applying random network coding to these constructed multiple trees, differently coded packets can be transmitted on multiple trees, and the throughput of the reliable multicast can be improved. Simulation results show that the average throughput can be enhanced 30% and the throughput of the poorly-connected receivers can be enhanced 100% with our protocol, compared to a recent work.
{"title":"Multiple Trees with Network Coding for Efficient and Reliable Multicast in MANETs","authors":"Yu-Hsun Chen, Gen-Huey Chen, E. Wu","doi":"10.1109/ICPPW.2010.85","DOIUrl":"https://doi.org/10.1109/ICPPW.2010.85","url":null,"abstract":"The innovative exploitation of network coding technology could bring great design advantages to MANETs. First, wireless links are inherently lossy due to channel fading or interference. Second, the delivery condition of the route from the source to each receiver in a multicast transmission might be significantly distinct. Trying to satisfy the reliability requirement for the poorly-connected receivers may affect the performance of the rest of the receivers. In this paper, we present a tree-based multicast protocol, which exploits the characteristics of network coding to provide efficient and reliable multicast. A metric, named Receiving Probability (RP), is proposed to represent the probability for a node in a multicast tree to successfully receive a packet sent by the multicast source. To mitigate the distinctions among the receivers, the proposed multicast protocol aims to construct multiple trees such that the RP of each receiver satisfies a predefined threshold. Besides, since seldom redundant packets are generated with random network coding, by applying random network coding to these constructed multiple trees, differently coded packets can be transmitted on multiple trees, and the throughput of the reliable multicast can be improved. Simulation results show that the average throughput can be enhanced 30% and the throughput of the poorly-connected receivers can be enhanced 100% with our protocol, compared to a recent work.","PeriodicalId":415472,"journal":{"name":"2010 39th International Conference on Parallel Processing Workshops","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115327011","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}
Large-scale High Performance Computing (HPC) systems continue to be designed and constructed to extend performance beyond Petascales using monolithic, cluster, and distributed architectures and emerging multi-core Central Processing Unit (CPU) technologies. As these machines grow so too does the size and variety of applications that run on them. Yet power management and interconnection performance are of great and mounting concern, and to date, the understanding of HPC subsystem interactions and their relationships to power efficiency remains less than desirable. Furthermore, Executive Order 13423 was issued in January of 2007 in an effort to ensure that Federal agencies operate in an environmentally, economically, and fiscally sound manner. It mandates a 30% reduction in energy intensity (MBTUs per square foot) of government facilities in the FY06-15 timeframe using FY03 as a baseline. Two major drawbacks hinder their ability to sustain consistent run time and energy efficient performance: (1) major subsystems interact with each other, often at the expense of unpredictable application run time and energy consumption, and (2) increased power density of these machines complicates the space, power, and cooling problem, resulting in partial or full system down time, further exacerbating run time unpredictability. We believe that one fundamental reason for the above limitations is the operational isolation of loosely coupled subsystems. While the development of subsystems in isolation has been the dominant model for decades, it is inherently unsuitable for ensuring consistent and sustainable systemic performance. We propose that the collection of HPC sub-systems, including the set of running applications must be collaborative in nature, and as such the HPC systems full potential is limited by subsystem isolation and autonomous actions to improve their individual subsystem performance. This paper describes an approach for using “Application-Level Behavioral Attribute Driven Techniques” to characterize HPC subsystem interactions into meaningful metrics and correlates that can be used as inputs to algorithms to control large-scale behaviors (job schedulers, routers, and HVAC systems) as well as smaller-scale behaviors such as CPU frequency and voltage scaling to achieve improved run time and energy efficiency to help satisfy Executive Order 13423.
{"title":"On Performance and Energy Management in High Performance Computing Systems","authors":"Jeffrey J. Evans","doi":"10.1109/ICPPW.2010.66","DOIUrl":"https://doi.org/10.1109/ICPPW.2010.66","url":null,"abstract":"Large-scale High Performance Computing (HPC) systems continue to be designed and constructed to extend performance beyond Petascales using monolithic, cluster, and distributed architectures and emerging multi-core Central Processing Unit (CPU) technologies. As these machines grow so too does the size and variety of applications that run on them. Yet power management and interconnection performance are of great and mounting concern, and to date, the understanding of HPC subsystem interactions and their relationships to power efficiency remains less than desirable. Furthermore, Executive Order 13423 was issued in January of 2007 in an effort to ensure that Federal agencies operate in an environmentally, economically, and fiscally sound manner. It mandates a 30% reduction in energy intensity (MBTUs per square foot) of government facilities in the FY06-15 timeframe using FY03 as a baseline. Two major drawbacks hinder their ability to sustain consistent run time and energy efficient performance: (1) major subsystems interact with each other, often at the expense of unpredictable application run time and energy consumption, and (2) increased power density of these machines complicates the space, power, and cooling problem, resulting in partial or full system down time, further exacerbating run time unpredictability. We believe that one fundamental reason for the above limitations is the operational isolation of loosely coupled subsystems. While the development of subsystems in isolation has been the dominant model for decades, it is inherently unsuitable for ensuring consistent and sustainable systemic performance. We propose that the collection of HPC sub-systems, including the set of running applications must be collaborative in nature, and as such the HPC systems full potential is limited by subsystem isolation and autonomous actions to improve their individual subsystem performance. This paper describes an approach for using “Application-Level Behavioral Attribute Driven Techniques” to characterize HPC subsystem interactions into meaningful metrics and correlates that can be used as inputs to algorithms to control large-scale behaviors (job schedulers, routers, and HVAC systems) as well as smaller-scale behaviors such as CPU frequency and voltage scaling to achieve improved run time and energy efficiency to help satisfy Executive Order 13423.","PeriodicalId":415472,"journal":{"name":"2010 39th International Conference on Parallel Processing Workshops","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128948181","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}
Electrocardiography (ECG) is an interpretation of the electrical activity of the heart over time captured and externally recorded by electrodes. An effective approach to speed up this and other biomedical operations is to integrate a very high number of processing elements in a single chip so that the massive scale of fine-grain parallelism inherent in several biomedical applications can be exploited efficiently. In this paper, we exploit parallel processing techniques to process electrocardiography computation kernels in parallel. We present an efficient ECG analysis algorithm based on Period-Peak Detection (PPD) approach. The system is implemented in a multicore System-on-Chip. System architecture and evaluation results are given in detail.
{"title":"An Efficient Algorithm and Embedded Multicore Implementation of ECG Analysis in Multi-lead Electrocardiogram Records","authors":"B. Abderazek, Y. Haga, Kenichi Kuroda","doi":"10.1109/ICPPW.2010.25","DOIUrl":"https://doi.org/10.1109/ICPPW.2010.25","url":null,"abstract":"Electrocardiography (ECG) is an interpretation of the electrical activity of the heart over time captured and externally recorded by electrodes. An effective approach to speed up this and other biomedical operations is to integrate a very high number of processing elements in a single chip so that the massive scale of fine-grain parallelism inherent in several biomedical applications can be exploited efficiently. In this paper, we exploit parallel processing techniques to process electrocardiography computation kernels in parallel. We present an efficient ECG analysis algorithm based on Period-Peak Detection (PPD) approach. The system is implemented in a multicore System-on-Chip. System architecture and evaluation results are given in detail.","PeriodicalId":415472,"journal":{"name":"2010 39th International Conference on Parallel Processing Workshops","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129814154","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：