Pub Date : 2015-12-17DOI: 10.1109/ESTIMedia.2015.7351772
N. Khalilzad, M. Ashjaei, L. Almeida, M. Behnam, Thomas Nolte
Complexity in the real-time embedded software domain has been growing rapidly. The component-based software development approach facilitates the development process of such software systems by dividing a complex system into a number of simpler components. Resource reservation techniques have been widely used for providing resources to real-time software components. In this paper we target real-time components operating on a distributed resource infrastructure. Furthermore, we target a class of software components which demonstrate dynamic resource consumption behavior. A prime example of such components is a multimedia software component. In the paper, we present a framework supporting multi-resource endto- end resource reservations. We reserve resource bandwidths on both processor resources as well as on the network resources. The proposed framework utilizes a Multiple Input Multiple Output (MIMO) controller which adjusts the sizes of reservations tracking the dynamic resource demands of the software components. Finally, we present a case study using a multimedia component to demonstrate the performance and efficiency of our framework.
{"title":"Adaptive multi-resource end-to-end reservations for component-based distributed real-time systems","authors":"N. Khalilzad, M. Ashjaei, L. Almeida, M. Behnam, Thomas Nolte","doi":"10.1109/ESTIMedia.2015.7351772","DOIUrl":"https://doi.org/10.1109/ESTIMedia.2015.7351772","url":null,"abstract":"Complexity in the real-time embedded software domain has been growing rapidly. The component-based software development approach facilitates the development process of such software systems by dividing a complex system into a number of simpler components. Resource reservation techniques have been widely used for providing resources to real-time software components. In this paper we target real-time components operating on a distributed resource infrastructure. Furthermore, we target a class of software components which demonstrate dynamic resource consumption behavior. A prime example of such components is a multimedia software component. In the paper, we present a framework supporting multi-resource endto- end resource reservations. We reserve resource bandwidths on both processor resources as well as on the network resources. The proposed framework utilizes a Multiple Input Multiple Output (MIMO) controller which adjusts the sizes of reservations tracking the dynamic resource demands of the software components. Finally, we present a case study using a multimedia component to demonstrate the performance and efficiency of our framework.","PeriodicalId":350361,"journal":{"name":"2015 13th IEEE Symposium on Embedded Systems For Real-time Multimedia (ESTIMedia)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122197165","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 : 2015-12-17DOI: 10.1109/ESTIMedia.2015.7351774
Siddharth Advani, Brigid Smith, Yasuki Tanabe, K. Irick, M. Cotter, J. Sampson, N. Vijaykrishnan
In any visual object recognition system, the classification accuracy will likely determine the usefulness of the system as a whole. In many real-world applications, it is also important to be able to recognize a large number of diverse objects for the system to be robust enough to handle the sort of tasks that the human visual system handles on an average day. These objectives are often at odds with performance, as running too large of a number of detectors on any one scene will be prohibitively slow for use in any real-time scenario. However, visual information has temporal and spatial context that can be exploited to reduce the number of detectors that need to be triggered at any given instance. In this paper, we propose a dynamic approach to encode such context, called Visual Co-occurrence Network (ViCoNet) that establishes relationships between objects observed in a visual scene. We investigate the utility of ViCoNet when integrated into a vision pipeline targeted for retail shopping. When evaluated on a large and deep dataset, we achieve a 50% improvement in performance and a 7% improvement in accuracy in the best case, and a 45% improvement in performance and a 3% improvement in accuracy in the average case over an established baseline. The memory overhead of ViCoNet is around 10KB, highlighting its effectiveness on temporal big data.
{"title":"Visual co-occurrence network: using context for large-scale object recognition in retail","authors":"Siddharth Advani, Brigid Smith, Yasuki Tanabe, K. Irick, M. Cotter, J. Sampson, N. Vijaykrishnan","doi":"10.1109/ESTIMedia.2015.7351774","DOIUrl":"https://doi.org/10.1109/ESTIMedia.2015.7351774","url":null,"abstract":"In any visual object recognition system, the classification accuracy will likely determine the usefulness of the system as a whole. In many real-world applications, it is also important to be able to recognize a large number of diverse objects for the system to be robust enough to handle the sort of tasks that the human visual system handles on an average day. These objectives are often at odds with performance, as running too large of a number of detectors on any one scene will be prohibitively slow for use in any real-time scenario. However, visual information has temporal and spatial context that can be exploited to reduce the number of detectors that need to be triggered at any given instance. In this paper, we propose a dynamic approach to encode such context, called Visual Co-occurrence Network (ViCoNet) that establishes relationships between objects observed in a visual scene. We investigate the utility of ViCoNet when integrated into a vision pipeline targeted for retail shopping. When evaluated on a large and deep dataset, we achieve a 50% improvement in performance and a 7% improvement in accuracy in the best case, and a 45% improvement in performance and a 3% improvement in accuracy in the average case over an established baseline. The memory overhead of ViCoNet is around 10KB, highlighting its effectiveness on temporal big data.","PeriodicalId":350361,"journal":{"name":"2015 13th IEEE Symposium on Embedded Systems For Real-time Multimedia (ESTIMedia)","volume":"215 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130866224","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 : 2015-12-17DOI: 10.1109/ESTIMedia.2015.7351768
Hyukwoo Park, Wonki Jung, Soo-Mook Moon
Web applications (apps) programmed using HTML, CSS, and JavaScript have advantages in portability and productivity, compared to Android or IOS apps. However, web apps are involved with some performance issue, due to JavaScript with its dynamic typing and prototypes which are difficult to execute efficiently. One popular way of accelerating JavaScript is using the just-in-time compilation (JITC), which translates the JavaScript source code to the machine code at runtime. Unfortunately, JavaScript JITC for web apps suffers from the parsing and compilation overhead seriously, which offsets the performance gain of executing the compiled code. In this paper, we propose ahead-of-time compilation (AOTC) of JavaScript at the client device. We save the code generated by the JITC at the first run of the web app, so that we can reuse the code in the next runs to remove the parsing and the compilation overhead. For the JavaScriptCore (JSC) engine of the WebKit, we developed three AOTCs. Bytecode-AOTC saves the bytecode, so we can omit the parsing overhead. Native-AOTC saves the machine code, so we can omit the compilation as well as the parsing overhead, yet with a higher space overhead. We also developed a selective-AOTC which selects between the two AOTC depending on the performance benefit and the space overhead of a target JavaScript function. We experimented with four web apps on a commercial smart TV. For the JavaScript portion of the app loading time, bytecode-AOTC and native- AOTC achieve 33.5% and 62.1% performance benefit, with a space overhead of 2.3 times and 15.4 times of the original JavaScript source code size, respectively. Selective-AOTC achieves a trade-off of 45.9% performance benefit and a space overhead of 4.7 times. Our evaluation on JavaScript benchmarks also shows a tangible performance gain, although it is lower than web apps. Finally, we compare to the AOTC for V8 engine and discuss some issues.
{"title":"Javascript ahead-of-time compilation for embedded web platform","authors":"Hyukwoo Park, Wonki Jung, Soo-Mook Moon","doi":"10.1109/ESTIMedia.2015.7351768","DOIUrl":"https://doi.org/10.1109/ESTIMedia.2015.7351768","url":null,"abstract":"Web applications (apps) programmed using HTML, CSS, and JavaScript have advantages in portability and productivity, compared to Android or IOS apps. However, web apps are involved with some performance issue, due to JavaScript with its dynamic typing and prototypes which are difficult to execute efficiently. One popular way of accelerating JavaScript is using the just-in-time compilation (JITC), which translates the JavaScript source code to the machine code at runtime. Unfortunately, JavaScript JITC for web apps suffers from the parsing and compilation overhead seriously, which offsets the performance gain of executing the compiled code. In this paper, we propose ahead-of-time compilation (AOTC) of JavaScript at the client device. We save the code generated by the JITC at the first run of the web app, so that we can reuse the code in the next runs to remove the parsing and the compilation overhead. For the JavaScriptCore (JSC) engine of the WebKit, we developed three AOTCs. Bytecode-AOTC saves the bytecode, so we can omit the parsing overhead. Native-AOTC saves the machine code, so we can omit the compilation as well as the parsing overhead, yet with a higher space overhead. We also developed a selective-AOTC which selects between the two AOTC depending on the performance benefit and the space overhead of a target JavaScript function. We experimented with four web apps on a commercial smart TV. For the JavaScript portion of the app loading time, bytecode-AOTC and native- AOTC achieve 33.5% and 62.1% performance benefit, with a space overhead of 2.3 times and 15.4 times of the original JavaScript source code size, respectively. Selective-AOTC achieves a trade-off of 45.9% performance benefit and a space overhead of 4.7 times. Our evaluation on JavaScript benchmarks also shows a tangible performance gain, although it is lower than web apps. Finally, we compare to the AOTC for V8 engine and discuss some issues.","PeriodicalId":350361,"journal":{"name":"2015 13th IEEE Symposium on Embedded Systems For Real-time Multimedia (ESTIMedia)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126182172","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 : 2015-12-17DOI: 10.1109/ESTIMedia.2015.7351764
Di Liu, J. Spasić, Gang Chen, T. Stefanov
In this paper, we propose a novel polynomial time algorithm, called Frequency Driven Mapping, to map real-time streaming applications specified as cyclo-static dataflow (CSDF) graphs onto a cluster heterogeneous MPSoC. The objective of our mapping approach is to reduce the energy consumption and guarantee latency and throughput constraints. The main novelty in our mapping algorithm is twofold: (1) By using hard-realtime scheduling of CSDF graphs, we propose an efficient way to determine a suitable processor type for each task in a CSDF graph, where the energy consumption is minimized and throughput and latency constraints are met; (2) According to an initial mapping derived by a first-fit-decreasing heuristic, we propose a remapping approach, where some tasks are remapped to unused clusters in order to further reduce the energy consumption of the system by cluster dynamic voltage/frequency scaling (DVFS). The experimental results show that the proposed algorithm finds more energy efficient mapping compared to existing approaches. The energy savings due to our proposed algorithm are up to 34%.
{"title":"Energy-efficient mapping of real-time streaming applications on cluster heterogeneous MPSoCs","authors":"Di Liu, J. Spasić, Gang Chen, T. Stefanov","doi":"10.1109/ESTIMedia.2015.7351764","DOIUrl":"https://doi.org/10.1109/ESTIMedia.2015.7351764","url":null,"abstract":"In this paper, we propose a novel polynomial time algorithm, called Frequency Driven Mapping, to map real-time streaming applications specified as cyclo-static dataflow (CSDF) graphs onto a cluster heterogeneous MPSoC. The objective of our mapping approach is to reduce the energy consumption and guarantee latency and throughput constraints. The main novelty in our mapping algorithm is twofold: (1) By using hard-realtime scheduling of CSDF graphs, we propose an efficient way to determine a suitable processor type for each task in a CSDF graph, where the energy consumption is minimized and throughput and latency constraints are met; (2) According to an initial mapping derived by a first-fit-decreasing heuristic, we propose a remapping approach, where some tasks are remapped to unused clusters in order to further reduce the energy consumption of the system by cluster dynamic voltage/frequency scaling (DVFS). The experimental results show that the proposed algorithm finds more energy efficient mapping compared to existing approaches. The energy savings due to our proposed algorithm are up to 34%.","PeriodicalId":350361,"journal":{"name":"2015 13th IEEE Symposium on Embedded Systems For Real-time Multimedia (ESTIMedia)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126559893","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}
Web applications (apps) are programs created by web technologies such as HTML, CSS, and JavaScript. Web apps can be executed on any platform that supports a web browser. Such portability allows an interesting user experience called app migration [2, 4, 5], which can save app's execution state information as a snapshot, transmit to another device, and continue its execution on the device. However, there is one feasibility issue in existing approaches because they save app's state without separating the JavaScript framework state, resulting in both time and size overhead. JavaScript framework such as jQuery is a library written in JavaScript to support web app development, popularly used in web apps. So an app's JavaScript heap is composed of app's objects and framework objects, possibly intermingled with their pointers, and the app code can modify framework objects. In this paper, we propose framework separated migration, which can separate the framework objects during snapshot creation so that snapshot does not contain framework objects, but does contain the changes made to them as well as the app objects. After transmission, the browser loads the JavaScript framework first, loads the snapshot, and restores the original state of framework and app objects to continue execution. With our approach, we could reduce 40% of total migration time, resulting increased feasibility in a real situation.
{"title":"Framework separated migration for web applications","authors":"Jin-woo Kwon, JinSeok Oh, InChang Jeong, Soo-Mook Moon","doi":"10.1109/ESTIMedia.2015.7351767","DOIUrl":"https://doi.org/10.1109/ESTIMedia.2015.7351767","url":null,"abstract":"Web applications (apps) are programs created by web technologies such as HTML, CSS, and JavaScript. Web apps can be executed on any platform that supports a web browser. Such portability allows an interesting user experience called app migration [2, 4, 5], which can save app's execution state information as a snapshot, transmit to another device, and continue its execution on the device. However, there is one feasibility issue in existing approaches because they save app's state without separating the JavaScript framework state, resulting in both time and size overhead. JavaScript framework such as jQuery is a library written in JavaScript to support web app development, popularly used in web apps. So an app's JavaScript heap is composed of app's objects and framework objects, possibly intermingled with their pointers, and the app code can modify framework objects. In this paper, we propose framework separated migration, which can separate the framework objects during snapshot creation so that snapshot does not contain framework objects, but does contain the changes made to them as well as the app objects. After transmission, the browser loads the JavaScript framework first, loads the snapshot, and restores the original state of framework and app objects to continue execution. With our approach, we could reduce 40% of total migration time, resulting increased feasibility in a real situation.","PeriodicalId":350361,"journal":{"name":"2015 13th IEEE Symposium on Embedded Systems For Real-time Multimedia (ESTIMedia)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123473735","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 : 2015-12-17DOI: 10.1109/ESTIMedia.2015.7351763
Lars Middendorf, C. Haubelt
Although modern graphics processing units (GPU) contain a large number of programmable shader cores, the focus on data parallelism and also the lack of efficient on-chip communication hinder the creation of custom graphics pipelines with arbitrary topologies. Based on the concept of stream rewriting, we propose a novel many-core architecture for graphics processing, which supports dynamic scheduling of recursively expandable task graphs and graphics pipelines. In particular, the tasks and their dependencies are encoded as a token stream, which is iteratively rewritten via pattern matching on multiple cores in parallel. The scalability of the proposed hardware architecture has been evaluated using an FPGA prototype.
{"title":"Dynamic task mapping of graphics processing applications on many-core architectures through stream rewriting","authors":"Lars Middendorf, C. Haubelt","doi":"10.1109/ESTIMedia.2015.7351763","DOIUrl":"https://doi.org/10.1109/ESTIMedia.2015.7351763","url":null,"abstract":"Although modern graphics processing units (GPU) contain a large number of programmable shader cores, the focus on data parallelism and also the lack of efficient on-chip communication hinder the creation of custom graphics pipelines with arbitrary topologies. Based on the concept of stream rewriting, we propose a novel many-core architecture for graphics processing, which supports dynamic scheduling of recursively expandable task graphs and graphics pipelines. In particular, the tasks and their dependencies are encoded as a token stream, which is iteratively rewritten via pattern matching on multiple cores in parallel. The scalability of the proposed hardware architecture has been evaluated using an FPGA prototype.","PeriodicalId":350361,"journal":{"name":"2015 13th IEEE Symposium on Embedded Systems For Real-time Multimedia (ESTIMedia)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126040917","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 : 2015-12-17DOI: 10.1109/ESTIMedia.2015.7351760
Johny Paul, W. Stechele
Multiprocessor System-on-Chip (MPSoC) designs offer a lot of computational power assembled in a compact design. The computing power of MPSoCs can be further augmented by adding heterogeneous processing elements, e.g. massively parallel processor arrays (MPPA) and specialized hardware with instruction-set extensions. However, the presence of multiple processing elements (PEs) with different characteristics raises issues related to programming and application mapping. The conventional approach used for programming heterogeneous MPSoCs results in a static mapping of various parts of the application to different PE types, based on the nature of the algorithm and the structure of the PEs. Yet, such a mapping scheme independent of the instantaneous load on the PEs may lead to under-utilization of some type of PEs while overloading others. We investigate the benefits of a resource-aware programming model called Invasive Computing for dynamically mapping image processing applications to different types of PEs available on a heterogeneous MPSoC. A case study of visual object recognition is presented, including Harris corner detection and SIFT feature matching. Results indicate that resource-aware programming helps to predict the latency of the application program along with better overall workload distribution within the heterogeneous MPSoC.
{"title":"Predictability of image processing algorithms on heterogeneous MPSoC","authors":"Johny Paul, W. Stechele","doi":"10.1109/ESTIMedia.2015.7351760","DOIUrl":"https://doi.org/10.1109/ESTIMedia.2015.7351760","url":null,"abstract":"Multiprocessor System-on-Chip (MPSoC) designs offer a lot of computational power assembled in a compact design. The computing power of MPSoCs can be further augmented by adding heterogeneous processing elements, e.g. massively parallel processor arrays (MPPA) and specialized hardware with instruction-set extensions. However, the presence of multiple processing elements (PEs) with different characteristics raises issues related to programming and application mapping. The conventional approach used for programming heterogeneous MPSoCs results in a static mapping of various parts of the application to different PE types, based on the nature of the algorithm and the structure of the PEs. Yet, such a mapping scheme independent of the instantaneous load on the PEs may lead to under-utilization of some type of PEs while overloading others. We investigate the benefits of a resource-aware programming model called Invasive Computing for dynamically mapping image processing applications to different types of PEs available on a heterogeneous MPSoC. A case study of visual object recognition is presented, including Harris corner detection and SIFT feature matching. Results indicate that resource-aware programming helps to predict the latency of the application program along with better overall workload distribution within the heterogeneous MPSoC.","PeriodicalId":350361,"journal":{"name":"2015 13th IEEE Symposium on Embedded Systems For Real-time Multimedia (ESTIMedia)","volume":"78 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127419033","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 : 2015-12-17DOI: 10.1109/ESTIMedia.2015.7351773
Junhyung Moon, Kyoungwoo Lee
Various video applications in mobile and wearable devices deal with private or important video data. In order to protect the important video information, several video encryption techniques have been proposed. The secure video processing, the combination of the video compression/decompression and the video encryption/decryption, causes lots of computational overheads, thereby consuming huge energy. Therefore, mobile and wearable devices which have limited battery capacity need to efficiently manage the power resources when performing the secure video processing. In order to find the configuration of the secure video processing which satisfies both the energy and visual security constraints, examining the tradeoff space in the secure video processing is significantly essential. In this paper, we study the tradeoff space between the energy consumption and the visual security in the secure video processing. Consequently, we propose the EVS (Energy-bound Visual Security) to find the interesting configuration of the secure video processing in order to achieve the maximum visual security under the energy budget, by exploring the huge tradeoff space. Moreover, we design an exploration technique to reduce the overhead of investigating the tradeoff space. Thanks to the EVS, we improve the visual security by up to about 17% under the same energy budget in our experiments. In addition, our proposed technique reduces about 55% exploration overhead in the experiments.
{"title":"Integrated visual security management for video encryption in limited battery devices","authors":"Junhyung Moon, Kyoungwoo Lee","doi":"10.1109/ESTIMedia.2015.7351773","DOIUrl":"https://doi.org/10.1109/ESTIMedia.2015.7351773","url":null,"abstract":"Various video applications in mobile and wearable devices deal with private or important video data. In order to protect the important video information, several video encryption techniques have been proposed. The secure video processing, the combination of the video compression/decompression and the video encryption/decryption, causes lots of computational overheads, thereby consuming huge energy. Therefore, mobile and wearable devices which have limited battery capacity need to efficiently manage the power resources when performing the secure video processing. In order to find the configuration of the secure video processing which satisfies both the energy and visual security constraints, examining the tradeoff space in the secure video processing is significantly essential. In this paper, we study the tradeoff space between the energy consumption and the visual security in the secure video processing. Consequently, we propose the EVS (Energy-bound Visual Security) to find the interesting configuration of the secure video processing in order to achieve the maximum visual security under the energy budget, by exploring the huge tradeoff space. Moreover, we design an exploration technique to reduce the overhead of investigating the tradeoff space. Thanks to the EVS, we improve the visual security by up to about 17% under the same energy budget in our experiments. In addition, our proposed technique reduces about 55% exploration overhead in the experiments.","PeriodicalId":350361,"journal":{"name":"2015 13th IEEE Symposium on Embedded Systems For Real-time Multimedia (ESTIMedia)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125307725","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 : 2015-12-17DOI: 10.1109/ESTIMedia.2015.7351761
Sascha Roloff, S. Wildermann, Frank Hannig, J. Teich
Heterogeneous many-core systems enable the integration of more and more applications into a single system. Executing multiple applications in the same system inevitably leads to resource sharing, e.g., when accessing on-chip communication and memory. This poses a challenge when applications are expected to guarantee user requirements regarding timing, reliability, security, etc. In this paper, we review a design methodology that (a) allows an application designer to model a stream processing application and user requirements, and then (b) automatically generates a set of resource requirements that guarantee the fulfillment of these user requirements. Techniques from the Invasive Computing paradigm enable the program-driven dynamic reservation of resources according to these generated resource requirements.We demonstrate that this provides means for predictable execution of stream processing applications by evaluating a simulation-based case study.
{"title":"Invasive computing for predictable stream processing: a simulation-based case study","authors":"Sascha Roloff, S. Wildermann, Frank Hannig, J. Teich","doi":"10.1109/ESTIMedia.2015.7351761","DOIUrl":"https://doi.org/10.1109/ESTIMedia.2015.7351761","url":null,"abstract":"Heterogeneous many-core systems enable the integration of more and more applications into a single system. Executing multiple applications in the same system inevitably leads to resource sharing, e.g., when accessing on-chip communication and memory. This poses a challenge when applications are expected to guarantee user requirements regarding timing, reliability, security, etc. In this paper, we review a design methodology that (a) allows an application designer to model a stream processing application and user requirements, and then (b) automatically generates a set of resource requirements that guarantee the fulfillment of these user requirements. Techniques from the Invasive Computing paradigm enable the program-driven dynamic reservation of resources according to these generated resource requirements.We demonstrate that this provides means for predictable execution of stream processing applications by evaluating a simulation-based case study.","PeriodicalId":350361,"journal":{"name":"2015 13th IEEE Symposium on Embedded Systems For Real-time Multimedia (ESTIMedia)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134452677","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 : 2015-12-17DOI: 10.1109/ESTIMedia.2015.7351769
Sajjad Taheri, Laleh Aghababaie Beni, A. Veidenbaum, A. Nicolau, Rosario Cammarota, Jianlin Qiu, Qiang Lu, M. Haghighat
WebRTC is an HTML5 API that allows browsers to establish a peer-to-peer connection for transferring data and media content via JavaScript APIs. This functionality enables broad range of new applications to emerge and is going to revolutionize Web communication. However, this technology is still under development and standardization process. Hence, detecting performance bottlenecks of different implementations across operating systems and architectures can help improve it significantly, and a benchmark suite would be a great help to accomplish this task. In this paper, we present WebRTCBench, a benchmark which measures WebRTC peer connection establishment and communication performance. We present and discuss performance evaluation of WebRTC implementations across a range of implementations and devices. This benchmark is publicly available under GPL license.
{"title":"WebRTCbench: a benchmark for performance assessment of webRTC implementations","authors":"Sajjad Taheri, Laleh Aghababaie Beni, A. Veidenbaum, A. Nicolau, Rosario Cammarota, Jianlin Qiu, Qiang Lu, M. Haghighat","doi":"10.1109/ESTIMedia.2015.7351769","DOIUrl":"https://doi.org/10.1109/ESTIMedia.2015.7351769","url":null,"abstract":"WebRTC is an HTML5 API that allows browsers to establish a peer-to-peer connection for transferring data and media content via JavaScript APIs. This functionality enables broad range of new applications to emerge and is going to revolutionize Web communication. However, this technology is still under development and standardization process. Hence, detecting performance bottlenecks of different implementations across operating systems and architectures can help improve it significantly, and a benchmark suite would be a great help to accomplish this task. In this paper, we present WebRTCBench, a benchmark which measures WebRTC peer connection establishment and communication performance. We present and discuss performance evaluation of WebRTC implementations across a range of implementations and devices. This benchmark is publicly available under GPL license.","PeriodicalId":350361,"journal":{"name":"2015 13th IEEE Symposium on Embedded Systems For Real-time Multimedia (ESTIMedia)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128290514","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：