Compared with rapid technology advancements in multicore processors and rapid changes from computing-intensive to highly data-intensive applications, operating systems have been evolved very slowly for several decades. Application users are facing to two major challenges in today's computing environment. On the top level of the system hierarchy, private and shared caches are equipped for many cores to access concurrently, inevitably causing access conflicts to degrade execution performance. On the bottom level, the performance bottleneck of "memory wall" has been shifted to "disk wall" that is a serious bottleneck for many data- intensive applications. Since processor caches and disk storage are not in the major scope of operating system management, and their increasingly complex operations are not transparent to application users, the above mentioned performance issues have not been effectively addressed at any level of computer systems. We have made a continuous effort to enhance operating systems with two objectives: (1) to well utilize rich but complex resources of multicore processors and (2) to access disk data as fast as possible. At the multicore processor level, we are developing new resource allocation management to improve the effective caching capacity per core and/or per thread, and to minimize congestion in off-chip memory accesses by coordinating memory bandwidth sharing. At the storage level, we enable operating systems to effectively exploit "sequential locality" - for the same amount of data, sequential accesses are several orders of magnitude faster than random accesses in disks. In this talk, related research issues and challenges will be overviewed, and preliminary results will be presented.
{"title":"Research Issues and Challenges to Advance System Software for Multicore Processors and Data-Intensive Applications","authors":"Xiaodong Zhang","doi":"10.1109/EUC.2008.189","DOIUrl":"https://doi.org/10.1109/EUC.2008.189","url":null,"abstract":"Compared with rapid technology advancements in multicore processors and rapid changes from computing-intensive to highly data-intensive applications, operating systems have been evolved very slowly for several decades. Application users are facing to two major challenges in today's computing environment. On the top level of the system hierarchy, private and shared caches are equipped for many cores to access concurrently, inevitably causing access conflicts to degrade execution performance. On the bottom level, the performance bottleneck of \"memory wall\" has been shifted to \"disk wall\" that is a serious bottleneck for many data- intensive applications. Since processor caches and disk storage are not in the major scope of operating system management, and their increasingly complex operations are not transparent to application users, the above mentioned performance issues have not been effectively addressed at any level of computer systems. We have made a continuous effort to enhance operating systems with two objectives: (1) to well utilize rich but complex resources of multicore processors and (2) to access disk data as fast as possible. At the multicore processor level, we are developing new resource allocation management to improve the effective caching capacity per core and/or per thread, and to minimize congestion in off-chip memory accesses by coordinating memory bandwidth sharing. At the storage level, we enable operating systems to effectively exploit \"sequential locality\" - for the same amount of data, sequential accesses are several orders of magnitude faster than random accesses in disks. In this talk, related research issues and challenges will be overviewed, and preliminary results will be presented.","PeriodicalId":430277,"journal":{"name":"2008 IEEE/IFIP International Conference on Embedded and Ubiquitous Computing","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128421449","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 aim of human activity recognition is to identify what a user or a group of users are doing at a given point in time, for example travelling or working. Activity recognition plays an important role in mobile and ubiquitous computing both as a goal in itself and as an intermediate task in the design of advanced applications. Virtually all existing activity recognition systems for mobile phones base their predictions on location cues. This approach forces the user to disclose personal information such as her home or work area. In this paper, we present a novel activity recognition system called TRAcME (temporal recognition of activities for mobile environments) which recognises generic human activities from large windows of context, Allenpsilas temporal relations and anonymous landmarks. Unlike existing systems, TRAcME handles simultaneous activities and outputs activities which are consistent with each other at the scale of a userpsilas day.
{"title":"TRAcME: Temporal Activity Recognition Using Mobile Phone Data","authors":"Driss Choujaa, Naranker Dulay","doi":"10.1109/EUC.2008.33","DOIUrl":"https://doi.org/10.1109/EUC.2008.33","url":null,"abstract":"The aim of human activity recognition is to identify what a user or a group of users are doing at a given point in time, for example travelling or working. Activity recognition plays an important role in mobile and ubiquitous computing both as a goal in itself and as an intermediate task in the design of advanced applications. Virtually all existing activity recognition systems for mobile phones base their predictions on location cues. This approach forces the user to disclose personal information such as her home or work area. In this paper, we present a novel activity recognition system called TRAcME (temporal recognition of activities for mobile environments) which recognises generic human activities from large windows of context, Allenpsilas temporal relations and anonymous landmarks. Unlike existing systems, TRAcME handles simultaneous activities and outputs activities which are consistent with each other at the scale of a userpsilas day.","PeriodicalId":430277,"journal":{"name":"2008 IEEE/IFIP International Conference on Embedded and Ubiquitous Computing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131968589","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}
Ambient Networks enable co-operation between heterogeneous networks, on demand, in order to provide ubiquitous connectivity to end-users. We focus on the part of disseminating context information to enable context-based applications (for e.g., resource update and management) and systems in ambient and ubiquitous networks. Our pure overlay space architecture acts as a base for the development and maintenance of application-layer context-aware dissemination protocols. Our multi-level overlay network based dissemination protocol exhibits dynamic and fair context dissemination. In this paper, we first overview this 'pure overlay' and 'multi-level overlay' combined context dissemination scheme and discuss its constraints in autonomous and heterogeneous environments. We then focus on proposing techniques for improving this dissemination scheme towards ubiquitous networking which satisfies our goals with regard to creating a user-friendly data model. The proposed optimizations include: local transformations, personalized extensions, and hybrid customizations. This paper is complemented with case study evaluation and simulation study results.
{"title":"CSON-D: An Ambient Network Aware Context Dissemination Scheme towards Ubiquitous Networking","authors":"D. Balakrishnan, A. Nayak, Pulak Dhar","doi":"10.1109/EUC.2008.145","DOIUrl":"https://doi.org/10.1109/EUC.2008.145","url":null,"abstract":"Ambient Networks enable co-operation between heterogeneous networks, on demand, in order to provide ubiquitous connectivity to end-users. We focus on the part of disseminating context information to enable context-based applications (for e.g., resource update and management) and systems in ambient and ubiquitous networks. Our pure overlay space architecture acts as a base for the development and maintenance of application-layer context-aware dissemination protocols. Our multi-level overlay network based dissemination protocol exhibits dynamic and fair context dissemination. In this paper, we first overview this 'pure overlay' and 'multi-level overlay' combined context dissemination scheme and discuss its constraints in autonomous and heterogeneous environments. We then focus on proposing techniques for improving this dissemination scheme towards ubiquitous networking which satisfies our goals with regard to creating a user-friendly data model. The proposed optimizations include: local transformations, personalized extensions, and hybrid customizations. This paper is complemented with case study evaluation and simulation study results.","PeriodicalId":430277,"journal":{"name":"2008 IEEE/IFIP International Conference on Embedded and Ubiquitous Computing","volume":"1155 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134472154","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}
In this paper, we present RT-P2P, a real-time peer-to-peer (P2P) system that allows application-level end-to-end timing requirements to be satisfied in P2P systems. Key aspects of our RT-P2P infrastructure include a real-time P2P protocol, real-time communication algorithm, and analytical performance models. We analytically establish the timing properties of RT-P2P. Our simulation studies validate the analytical results and demonstrate RT-P2P outperforms the traditional client-server model in large-scale and dynamic system.
{"title":"RT-P2P: A Scalable Real-Time Peer-to-Peer System with Probabilistic Timing Assurances","authors":"Fei Huang, B. Ravindran, E. Jensen","doi":"10.1109/EUC.2008.24","DOIUrl":"https://doi.org/10.1109/EUC.2008.24","url":null,"abstract":"In this paper, we present RT-P2P, a real-time peer-to-peer (P2P) system that allows application-level end-to-end timing requirements to be satisfied in P2P systems. Key aspects of our RT-P2P infrastructure include a real-time P2P protocol, real-time communication algorithm, and analytical performance models. We analytically establish the timing properties of RT-P2P. Our simulation studies validate the analytical results and demonstrate RT-P2P outperforms the traditional client-server model in large-scale and dynamic system.","PeriodicalId":430277,"journal":{"name":"2008 IEEE/IFIP International Conference on Embedded and Ubiquitous Computing","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131786703","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 integration of heterogeneous intellectual property hardware (IP) into existing hardware platforms is often only considered with respect to IP cores that provide bus interfaces to their environment. This paper will show that there is a need for, and an advantage to be gained from, incorporating non-bus interface IP so that they can be used by software running on the platformpsilas processor(s). This paper presents a grammar with which services of non-bus interface cores can be described. The paper will show how this description can be used to generate adapters to integrate the IP as well as device drivers to use the IP. Experiments show how the use of this description can reduce the integration effort of non-bus interface cores.
{"title":"Automatic Integration of Non-Bus Hardware IP into SoC-Platforms for Use by Software","authors":"Robert Günzel","doi":"10.1109/EUC.2008.140","DOIUrl":"https://doi.org/10.1109/EUC.2008.140","url":null,"abstract":"The integration of heterogeneous intellectual property hardware (IP) into existing hardware platforms is often only considered with respect to IP cores that provide bus interfaces to their environment. This paper will show that there is a need for, and an advantage to be gained from, incorporating non-bus interface IP so that they can be used by software running on the platformpsilas processor(s). This paper presents a grammar with which services of non-bus interface cores can be described. The paper will show how this description can be used to generate adapters to integrate the IP as well as device drivers to use the IP. Experiments show how the use of this description can reduce the integration effort of non-bus interface cores.","PeriodicalId":430277,"journal":{"name":"2008 IEEE/IFIP International Conference on Embedded and Ubiquitous Computing","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113979015","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}
To formally and precisely capture the security properties that access control should adhere to, access control models are usually written to bridge the rather wide gap in abstraction between policies and mechanisms. In this paper, we propose a new general approach for property verification for access control models. The approach defines a standardized structure for access control models, providing for both property verification and automated generation of test cases. The approach expresses access control models in the specification language of a model checker and expresses generic access control properties in the property language. Then the approach uses the model checker to verify these properties for the access control models and generates test cases via combinatorial covering array for the system implementations of the models.
{"title":"Property Verification for Generic Access Control Models","authors":"Vincent C. Hu, D. R. Kuhn, Tao Xie","doi":"10.1109/EUC.2008.22","DOIUrl":"https://doi.org/10.1109/EUC.2008.22","url":null,"abstract":"To formally and precisely capture the security properties that access control should adhere to, access control models are usually written to bridge the rather wide gap in abstraction between policies and mechanisms. In this paper, we propose a new general approach for property verification for access control models. The approach defines a standardized structure for access control models, providing for both property verification and automated generation of test cases. The approach expresses access control models in the specification language of a model checker and expresses generic access control properties in the property language. Then the approach uses the model checker to verify these properties for the access control models and generates test cases via combinatorial covering array for the system implementations of the models.","PeriodicalId":430277,"journal":{"name":"2008 IEEE/IFIP International Conference on Embedded and Ubiquitous Computing","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124054991","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}
In a real-time Linux system, the critical sections are known as the main factor delaying the execution of real-time tasks. Traditional approaches to overcoming this issue have given less consideration to both real-time and non-real-time tasks. In this paper, we propose a new lock management mechanism to improve the real-time performance with a small penalty for non-real-time tasks. Using this mechanism, we guarantee the deadlines of real-time tasks while keeping the penalties accruing for non-real-time tasks small. We implemented a prototype system in Linux 2.6.20. Experimental results showed that the worst-case OS latency of real-time task is reduced to 19% of the original one, while the penalty for a non-real-time task is 10.1% of the original. The results also showed that the lock management mechanism proposed in this paper is efficient and useful for a future real-time Linux system.
{"title":"A Light Lock Management Mechanism for Optimizing Real-Time and Non-Real-Time Performance in Embedded Linux","authors":"Maobing Dai, T. Matsui, Y. Ishikawa","doi":"10.1109/EUC.2008.56","DOIUrl":"https://doi.org/10.1109/EUC.2008.56","url":null,"abstract":"In a real-time Linux system, the critical sections are known as the main factor delaying the execution of real-time tasks. Traditional approaches to overcoming this issue have given less consideration to both real-time and non-real-time tasks. In this paper, we propose a new lock management mechanism to improve the real-time performance with a small penalty for non-real-time tasks. Using this mechanism, we guarantee the deadlines of real-time tasks while keeping the penalties accruing for non-real-time tasks small. We implemented a prototype system in Linux 2.6.20. Experimental results showed that the worst-case OS latency of real-time task is reduced to 19% of the original one, while the penalty for a non-real-time task is 10.1% of the original. The results also showed that the lock management mechanism proposed in this paper is efficient and useful for a future real-time Linux system.","PeriodicalId":430277,"journal":{"name":"2008 IEEE/IFIP International Conference on Embedded and Ubiquitous Computing","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121389925","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}
Virtualization technology becomes more and more important in area of compute science, such as data center and server consolidation. A large number of hypervisors are available to manage the virtualization either on bare hardware or on host operating systems. One of the important task for the designer is to measure and compare the performance overhead of given virtual machines. In this paper, we provide an analytic framework for the performance analyzing either without running a system or in a runnable real system. Meanwhile, analytic performance models that are based on the queue network theory are developed to study the designs of virtual machines. At the end, a case study of the mathematical models is given to illustrate the performance evaluation.
{"title":"A Framework to Evaluate and Predict Performances in Virtual Machines Environment","authors":"Deshi Ye, Qinming He, Hua Chen, Jianhua Che","doi":"10.1109/EUC.2008.42","DOIUrl":"https://doi.org/10.1109/EUC.2008.42","url":null,"abstract":"Virtualization technology becomes more and more important in area of compute science, such as data center and server consolidation. A large number of hypervisors are available to manage the virtualization either on bare hardware or on host operating systems. One of the important task for the designer is to measure and compare the performance overhead of given virtual machines. In this paper, we provide an analytic framework for the performance analyzing either without running a system or in a runnable real system. Meanwhile, analytic performance models that are based on the queue network theory are developed to study the designs of virtual machines. At the end, a case study of the mathematical models is given to illustrate the performance evaluation.","PeriodicalId":430277,"journal":{"name":"2008 IEEE/IFIP International Conference on Embedded and Ubiquitous Computing","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116889667","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}
Virtual servers provide an abstraction layer to exploit the heterogeneity of peers (or nodes) that participate in a peer-to-peer (P2P) network based on the distributed hash table (DHT). In this paper, we present a load-balanced DHT network based on virtual servers. In our proposal, participating peers accept loads proportional to their capacities. By sampling the capacities of peers and workloads of virtual servers in the system, our design approximates the probability distributions for the capacities and workloads. Each node is based on the probability distributions to match their local virtual servers and other peers. Our proposal is developed through rigorous performance analysis, and is evaluated in extensive simulations.
{"title":"Load Balancing in Peer-to-Peer Networks Based on Sampling System States","authors":"Hung-Chang Hsiao, I-Ta Lin, Hao Liao","doi":"10.1109/EUC.2008.12","DOIUrl":"https://doi.org/10.1109/EUC.2008.12","url":null,"abstract":"Virtual servers provide an abstraction layer to exploit the heterogeneity of peers (or nodes) that participate in a peer-to-peer (P2P) network based on the distributed hash table (DHT). In this paper, we present a load-balanced DHT network based on virtual servers. In our proposal, participating peers accept loads proportional to their capacities. By sampling the capacities of peers and workloads of virtual servers in the system, our design approximates the probability distributions for the capacities and workloads. Each node is based on the probability distributions to match their local virtual servers and other peers. Our proposal is developed through rigorous performance analysis, and is evaluated in extensive simulations.","PeriodicalId":430277,"journal":{"name":"2008 IEEE/IFIP International Conference on Embedded and Ubiquitous Computing","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117300639","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}
Workflows in the physical world are implemented based on forms printed on paper and signed by the involved people. In contrast workflows in service oriented architectures require a central entity for execution that results in an inflexible system design that is not suitable for ambient environments. Based on an ambient environment example a security approach for decentralized workflows is presented providing authenticity, confidentiality, and non-repudiation for distributed workflow execution.
{"title":"Securing Decentralized Workflows in Ambient Environments","authors":"N. Kuntze, J. Schütte","doi":"10.1109/EUC.2008.86","DOIUrl":"https://doi.org/10.1109/EUC.2008.86","url":null,"abstract":"Workflows in the physical world are implemented based on forms printed on paper and signed by the involved people. In contrast workflows in service oriented architectures require a central entity for execution that results in an inflexible system design that is not suitable for ambient environments. Based on an ambient environment example a security approach for decentralized workflows is presented providing authenticity, confidentiality, and non-repudiation for distributed workflow execution.","PeriodicalId":430277,"journal":{"name":"2008 IEEE/IFIP International Conference on Embedded and Ubiquitous Computing","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116269641","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：