Janick Edinger, Dominik Schäfer, Martin Breitbach, C. Becker
{"title":"使用微线程开发分布式计算应用","authors":"Janick Edinger, Dominik Schäfer, Martin Breitbach, C. Becker","doi":"10.1109/PERCOMW.2017.7917534","DOIUrl":null,"url":null,"abstract":"This demo paper presents the Tasklet system - a middleware for distributed computing. The Tasklet system allows developers to offload self-contained units of computation - the so-called Tasklets - to a pool of heterogeneous computing devices. In this demonstration of the Tasklet system, we uncover the otherwise transparent process of computation offloading and scheduling. Further, we show the easy integration of heterogeneous computing devices and applications for Tasklets. In the demonstration, we focus on three key aspects of the Tasklet system. First, we present the API of the Tasklet middleware and the process of application tailoring with quality of computation. Second, we integrate computation devices during the runtime of Tasklets. A monitoring tool visualizes how Tasklets are scheduled and how resource providers simultaneously enter and leave the system. To show the easy integration of resources, we invite the audience to spontaneously contribute their computational resources to the environment. Third, we show five example applications from different application domains that benefit from using Tasklets.","PeriodicalId":319638,"journal":{"name":"2017 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":"{\"title\":\"Developing distributed computing applications with Tasklets\",\"authors\":\"Janick Edinger, Dominik Schäfer, Martin Breitbach, C. Becker\",\"doi\":\"10.1109/PERCOMW.2017.7917534\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This demo paper presents the Tasklet system - a middleware for distributed computing. The Tasklet system allows developers to offload self-contained units of computation - the so-called Tasklets - to a pool of heterogeneous computing devices. In this demonstration of the Tasklet system, we uncover the otherwise transparent process of computation offloading and scheduling. Further, we show the easy integration of heterogeneous computing devices and applications for Tasklets. In the demonstration, we focus on three key aspects of the Tasklet system. First, we present the API of the Tasklet middleware and the process of application tailoring with quality of computation. Second, we integrate computation devices during the runtime of Tasklets. A monitoring tool visualizes how Tasklets are scheduled and how resource providers simultaneously enter and leave the system. To show the easy integration of resources, we invite the audience to spontaneously contribute their computational resources to the environment. Third, we show five example applications from different application domains that benefit from using Tasklets.\",\"PeriodicalId\":319638,\"journal\":{\"name\":\"2017 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops)\",\"volume\":\"6 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-03-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"7\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2017 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/PERCOMW.2017.7917534\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/PERCOMW.2017.7917534","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Developing distributed computing applications with Tasklets
This demo paper presents the Tasklet system - a middleware for distributed computing. The Tasklet system allows developers to offload self-contained units of computation - the so-called Tasklets - to a pool of heterogeneous computing devices. In this demonstration of the Tasklet system, we uncover the otherwise transparent process of computation offloading and scheduling. Further, we show the easy integration of heterogeneous computing devices and applications for Tasklets. In the demonstration, we focus on three key aspects of the Tasklet system. First, we present the API of the Tasklet middleware and the process of application tailoring with quality of computation. Second, we integrate computation devices during the runtime of Tasklets. A monitoring tool visualizes how Tasklets are scheduled and how resource providers simultaneously enter and leave the system. To show the easy integration of resources, we invite the audience to spontaneously contribute their computational resources to the environment. Third, we show five example applications from different application domains that benefit from using Tasklets.