V. Kalogeraki, D. Gunopulos, R. Sandhu, B. Thuraisingham
In this paper we describe our approach for developing a QoS-aware, dependable execution environment for large-scale distributed stream processing applications. Distributed stream processing applications have strong timeliness and security demands. In particular, we address the following challenges: (1) propose a real-time dependable execution model by extending the component-based execution model with real-time and dependability properties, and (2) develop QoS-aware application composition and adaptation techniques that employ resource management strategies and security policies when discovering and selecting application components. Our approach enables us to develop a distributed stream processing environment that is predictable, secure, flexible and adaptable.
{"title":"QoS Aware Dependable Distributed Stream Processing","authors":"V. Kalogeraki, D. Gunopulos, R. Sandhu, B. Thuraisingham","doi":"10.1109/ISORC.2008.77","DOIUrl":"https://doi.org/10.1109/ISORC.2008.77","url":null,"abstract":"In this paper we describe our approach for developing a QoS-aware, dependable execution environment for large-scale distributed stream processing applications. Distributed stream processing applications have strong timeliness and security demands. In particular, we address the following challenges: (1) propose a real-time dependable execution model by extending the component-based execution model with real-time and dependability properties, and (2) develop QoS-aware application composition and adaptation techniques that employ resource management strategies and security policies when discovering and selecting application components. Our approach enables us to develop a distributed stream processing environment that is predictable, secure, flexible and adaptable.","PeriodicalId":378715,"journal":{"name":"2008 11th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing (ISORC)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115104893","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}
Computation, as expressed in modern programming languages, obscures many resource management problems. Memory is provided without bound by stacks and heaps. Power and energy consumption are not the concern of a programmer. Even when these resource management problems are important, there is no way to talk about them within the semantics of a programming language. Time, however, is not quite like these other resources. First, barring metaphysical discourse, it is genuinely unbounded. To say that "the available time per unit time is bounded" is tautological, yet this is effectively what people say when they manage it as a bounded resource. Second, time gets expended whether we use it or not. It cannot be conserved and saved for later. This is true up to a point with, say, battery power. Batteries leak, so their power cannot be indefinitely conserved, but designers rarely optimize a system to use as much battery power before it leaks away as they can. Yet that is what they do with time.
{"title":"Time is a Resource, and Other Stories","authors":"Edward A. Lee","doi":"10.1109/ISORC.2008.26","DOIUrl":"https://doi.org/10.1109/ISORC.2008.26","url":null,"abstract":"Computation, as expressed in modern programming languages, obscures many resource management problems. Memory is provided without bound by stacks and heaps. Power and energy consumption are not the concern of a programmer. Even when these resource management problems are important, there is no way to talk about them within the semantics of a programming language. Time, however, is not quite like these other resources. First, barring metaphysical discourse, it is genuinely unbounded. To say that \"the available time per unit time is bounded\" is tautological, yet this is effectively what people say when they manage it as a bounded resource. Second, time gets expended whether we use it or not. It cannot be conserved and saved for later. This is true up to a point with, say, battery power. Batteries leak, so their power cannot be indefinitely conserved, but designers rarely optimize a system to use as much battery power before it leaks away as they can. Yet that is what they do with time.","PeriodicalId":378715,"journal":{"name":"2008 11th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing (ISORC)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116805862","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}
Our aim is to investigate if it is possible to control the throughput (IPC rate) of a thread running on a simultaneous multi-threaded microprocessor by a closed feedback loop. We showed in previous experimental studies that the proposed approach works. In this paper we discuss the control theory approach from a theoretical point of view. We develop a mathematical model of a general purpose multi-threaded microprocessor enhanced with a closed feedback controller and use control theory methods to investigate properties like stability and settling time.
{"title":"A Control Theory Approach to Improve the Real-Time Capability of Multi-Threaded Microprocessors","authors":"U. Brinkschulte, Mathias Pacher","doi":"10.1109/ISORC.2008.8","DOIUrl":"https://doi.org/10.1109/ISORC.2008.8","url":null,"abstract":"Our aim is to investigate if it is possible to control the throughput (IPC rate) of a thread running on a simultaneous multi-threaded microprocessor by a closed feedback loop. We showed in previous experimental studies that the proposed approach works. In this paper we discuss the control theory approach from a theoretical point of view. We develop a mathematical model of a general purpose multi-threaded microprocessor enhanced with a closed feedback controller and use control theory methods to investigate properties like stability and settling time.","PeriodicalId":378715,"journal":{"name":"2008 11th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing (ISORC)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129551756","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 analysis of the worst-case execution time (WCET) requires detailed knowledge of the program behavior. In practice it is still not possible to obtain all needed information automatically. In this paper we present the current state of the art of WCET analysis and point to the main problems to be solved. The most eminent problem is the state problem, i.e., the precise determination of possible processor states at different program locations. The path problem refers to the fact that current tools are not able to calculate all (in)feasible paths automatically. We discuss how the main open problems manifest themselves in static and in measurement-based WCET analysis methods.
{"title":"Obstacles in Worst-Case Execution Time Analysis","authors":"R. Kirner, P. Puschner","doi":"10.1109/ISORC.2008.65","DOIUrl":"https://doi.org/10.1109/ISORC.2008.65","url":null,"abstract":"The analysis of the worst-case execution time (WCET) requires detailed knowledge of the program behavior. In practice it is still not possible to obtain all needed information automatically. In this paper we present the current state of the art of WCET analysis and point to the main problems to be solved. The most eminent problem is the state problem, i.e., the precise determination of possible processor states at different program locations. The path problem refers to the fact that current tools are not able to calculate all (in)feasible paths automatically. We discuss how the main open problems manifest themselves in static and in measurement-based WCET analysis methods.","PeriodicalId":378715,"journal":{"name":"2008 11th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing (ISORC)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129587335","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}
With society's increasing dependence on software, the issue of software quality is becoming more prominent. The term quality of service (QoS) has been used in conjunction with software to cover a broad range of characteristics, such as dependability, responsiveness, security, etc. Historically, the technical challenge of achieving a desired QoS in conditions of limited resources (e.g., finite computing power, memory capacity, communication bandwidth) was primarily relegated to the specialized domain of real-time and embedded software systems, that is, systems involved in continuous and timely interaction with the physical world. The "mainstream7" view and dominant software design philosophy are still based on the flawed premise that software should be developed without any considerations given to the characteristics and limitations of the hardware and software platforms that give it life. The important and useful principle of platform independence is being tragically misinterpreted as implying platform ignorance.
{"title":"Embedded Systems Research: Missed Opportunities","authors":"B. Selić","doi":"10.1109/ISORC.2008.78","DOIUrl":"https://doi.org/10.1109/ISORC.2008.78","url":null,"abstract":"With society's increasing dependence on software, the issue of software quality is becoming more prominent. The term quality of service (QoS) has been used in conjunction with software to cover a broad range of characteristics, such as dependability, responsiveness, security, etc. Historically, the technical challenge of achieving a desired QoS in conditions of limited resources (e.g., finite computing power, memory capacity, communication bandwidth) was primarily relegated to the specialized domain of real-time and embedded software systems, that is, systems involved in continuous and timely interaction with the physical world. The \"mainstream7\" view and dominant software design philosophy are still based on the flawed premise that software should be developed without any considerations given to the characteristics and limitations of the hardware and software platforms that give it life. The important and useful principle of platform independence is being tragically misinterpreted as implying platform ignorance.","PeriodicalId":378715,"journal":{"name":"2008 11th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing (ISORC)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114159357","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 time-triggered paradigm encompasses a set of concepts and principles that support the design of dependable real-time systems. By using the properties of physical time and the mechanism of clock synchronization, coordinated interaction between distributed nodes can be facilitated. This paper briefly explains the time-triggered approach, defines a taxonomy for real-time requirements and discusses seven protocols that provide time-triggered features. Finally, two application examples are given that show the potential of the time-triggered approach.
{"title":"Time-Triggered Fieldbus Networks State of the Art and Future Applications","authors":"W. Elmenreich","doi":"10.1109/ISORC.2008.50","DOIUrl":"https://doi.org/10.1109/ISORC.2008.50","url":null,"abstract":"The time-triggered paradigm encompasses a set of concepts and principles that support the design of dependable real-time systems. By using the properties of physical time and the mechanism of clock synchronization, coordinated interaction between distributed nodes can be facilitated. This paper briefly explains the time-triggered approach, defines a taxonomy for real-time requirements and discusses seven protocols that provide time-triggered features. Finally, two application examples are given that show the potential of the time-triggered approach.","PeriodicalId":378715,"journal":{"name":"2008 11th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing (ISORC)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116036174","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}
Xiaohua Dong, Ji Li, Zhongfu Wu, Dacheng Zhang, Jie Xu
Service oriented architecture (SOA) allows multiple and heterogeneous data resources to be integrated within a single service while hiding the implementation details and formats of data resources from users of the service. However, data sources for a service are often distributed geographically and connected with long-latency networks; time and bandwidth consumption of data transportation may have an impact on the system performance. Dynamic data replication is a practical solution to this problem. By replicating data copies to appropriate sites, this approach aims to reduce time and bandwidth consumptions over networks. Existing strategies for dynamic replication are typically based on so-called single-location algorithms for identifying a single site for data replication. In this paper we discuss the issues with single-location strategies in large-scale data integration applications, and examine potential multiple-location schemes. Dynamic multiple-location replication is NP-complete in nature. We therefore transform the multiple-location problem into several classical mathematical problems with different parameter settings, for which efficient approximation algorithms exist. Experimental results indicate that unlike single-location strategies our multiple-location schemes are efficient with respect to access latency and bandwidth consumption, especially when the requesters of a data set are distributed over a large scale of locations.
{"title":"On Dynamic Replication Strategies in Data Service Grids","authors":"Xiaohua Dong, Ji Li, Zhongfu Wu, Dacheng Zhang, Jie Xu","doi":"10.1109/ISORC.2008.66","DOIUrl":"https://doi.org/10.1109/ISORC.2008.66","url":null,"abstract":"Service oriented architecture (SOA) allows multiple and heterogeneous data resources to be integrated within a single service while hiding the implementation details and formats of data resources from users of the service. However, data sources for a service are often distributed geographically and connected with long-latency networks; time and bandwidth consumption of data transportation may have an impact on the system performance. Dynamic data replication is a practical solution to this problem. By replicating data copies to appropriate sites, this approach aims to reduce time and bandwidth consumptions over networks. Existing strategies for dynamic replication are typically based on so-called single-location algorithms for identifying a single site for data replication. In this paper we discuss the issues with single-location strategies in large-scale data integration applications, and examine potential multiple-location schemes. Dynamic multiple-location replication is NP-complete in nature. We therefore transform the multiple-location problem into several classical mathematical problems with different parameter settings, for which efficient approximation algorithms exist. Experimental results indicate that unlike single-location strategies our multiple-location schemes are efficient with respect to access latency and bandwidth consumption, especially when the requesters of a data set are distributed over a large scale of locations.","PeriodicalId":378715,"journal":{"name":"2008 11th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing (ISORC)","volume":"44 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120940585","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}
W. Tsai, Xin Sun, Yinong Chen, Qian Huang, G. Bitter, Mary White
This paper proposes a new approach to teach the STEM (Science, Technology, Engineering, and Mathematics) knowledge informally via robotic games. In this approach, a robotic playground is built to provide a hands-on programming and playing experience with robots controlled by Service-Oriented Computing (SOC) software, which is based on a new approach that uses reusable services (components) with standard interfaces and platform-independent interoperability. Services in the repository are annotated with STEM knowledge to enforce the required contents. In this way, students can learn computing and STEM in an entertaining manner.
{"title":"Teaching Service-Oriented Computing and STEM Topics via Robotic Games","authors":"W. Tsai, Xin Sun, Yinong Chen, Qian Huang, G. Bitter, Mary White","doi":"10.1109/ISORC.2008.83","DOIUrl":"https://doi.org/10.1109/ISORC.2008.83","url":null,"abstract":"This paper proposes a new approach to teach the STEM (Science, Technology, Engineering, and Mathematics) knowledge informally via robotic games. In this approach, a robotic playground is built to provide a hands-on programming and playing experience with robots controlled by Service-Oriented Computing (SOC) software, which is based on a new approach that uses reusable services (components) with standard interfaces and platform-independent interoperability. Services in the repository are annotated with STEM knowledge to enforce the required contents. In this way, students can learn computing and STEM in an entertaining manner.","PeriodicalId":378715,"journal":{"name":"2008 11th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing (ISORC)","volume":"241 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115204713","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}
An important part of implementing device drivers is to control the interrupt facilities of the hardware platform and to program interrupt handlers. Current methods for handling interrupts in Java use a server thread waiting for the VM to signal an interrupt occurrence. It means that the interrupt is handled at a later time, which has some disadvantages. We present constructs that allow interrupts to be handled directly and not at a later point decided by a scheduler. A desirable feature of our approach is that we do not require a native middelware layer but can handle interrupts entirely with Java code. We have implemented our approach using an interpreter and a Java processor, and give an example demonstrating its use.
{"title":"Interrupt Handlers in Java","authors":"Stephan Korsholm, Martin Schoeberl, A. Ravn","doi":"10.1109/ISORC.2008.68","DOIUrl":"https://doi.org/10.1109/ISORC.2008.68","url":null,"abstract":"An important part of implementing device drivers is to control the interrupt facilities of the hardware platform and to program interrupt handlers. Current methods for handling interrupts in Java use a server thread waiting for the VM to signal an interrupt occurrence. It means that the interrupt is handled at a later time, which has some disadvantages. We present constructs that allow interrupts to be handled directly and not at a later point decided by a scheduler. A desirable feature of our approach is that we do not require a native middelware layer but can handle interrupts entirely with Java code. We have implemented our approach using an interpreter and a Java processor, and give an example demonstrating its use.","PeriodicalId":378715,"journal":{"name":"2008 11th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing (ISORC)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123828139","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}
Jaegeuk Kim, Heeseung Jo, Hyotaek Shim, Jin-Soo Kim, S. Maeng
NAND flash memory becomes one of the most popular storage for portable embedded systems. Although many flash-aware file systems, such as JFFS2 and YAFFS2, were proposed, the large memory consumption and the long mount delay have been serious obstacles for large-capacity NAND flash memory. In this paper, we present a new flash-aware file system called DFFS (direct flash file system) which fetches only the needed metadata on demand from flash memory. In addition, DFFS employs two novel metadata management schemes, inode embedding scheme and hybrid inode indexing scheme, to improve the performance of metadata operations. Comprehensive evaluation results using microbench- mark, postmark, and Linux kernel compilation trace, show that DFFS has comparable performance to JFFS2 and YAFFS2, while achieving a small memory footprint and instant mount time.
{"title":"Efficient Metadata Management for Flash File Systems","authors":"Jaegeuk Kim, Heeseung Jo, Hyotaek Shim, Jin-Soo Kim, S. Maeng","doi":"10.1109/ISORC.2008.34","DOIUrl":"https://doi.org/10.1109/ISORC.2008.34","url":null,"abstract":"NAND flash memory becomes one of the most popular storage for portable embedded systems. Although many flash-aware file systems, such as JFFS2 and YAFFS2, were proposed, the large memory consumption and the long mount delay have been serious obstacles for large-capacity NAND flash memory. In this paper, we present a new flash-aware file system called DFFS (direct flash file system) which fetches only the needed metadata on demand from flash memory. In addition, DFFS employs two novel metadata management schemes, inode embedding scheme and hybrid inode indexing scheme, to improve the performance of metadata operations. Comprehensive evaluation results using microbench- mark, postmark, and Linux kernel compilation trace, show that DFFS has comparable performance to JFFS2 and YAFFS2, while achieving a small memory footprint and instant mount time.","PeriodicalId":378715,"journal":{"name":"2008 11th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing (ISORC)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129311864","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：