{"title":"开放社区运行时任务图的可视化","authors":"J. Dokulil, J. Katreniaková","doi":"10.1109/iV.2017.31","DOIUrl":null,"url":null,"abstract":"The emergence of new types of high performance hardware also drives the need for new programming models. The Open Community Runtime (OCR) proposal uses a task-based programming model to target some of these architectures. In OCR, the whole program from start to end needs to be expressed using tasks and synchronized using task-to-task dependences, significantly limiting the applicability and usefulness of existing approaches to application development and debugging. In this paper, we present our approach to visualizing tasks and their synchronization, based on trace data from application execution. This way, the application developer may compare the intended organization of the tasks with the actual dependences as they are seen by the OCR runtime system.","PeriodicalId":410876,"journal":{"name":"2017 21st International Conference Information Visualisation (IV)","volume":"102 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Visualization of Open Community Runtime Task Graphs\",\"authors\":\"J. Dokulil, J. Katreniaková\",\"doi\":\"10.1109/iV.2017.31\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The emergence of new types of high performance hardware also drives the need for new programming models. The Open Community Runtime (OCR) proposal uses a task-based programming model to target some of these architectures. In OCR, the whole program from start to end needs to be expressed using tasks and synchronized using task-to-task dependences, significantly limiting the applicability and usefulness of existing approaches to application development and debugging. In this paper, we present our approach to visualizing tasks and their synchronization, based on trace data from application execution. This way, the application developer may compare the intended organization of the tasks with the actual dependences as they are seen by the OCR runtime system.\",\"PeriodicalId\":410876,\"journal\":{\"name\":\"2017 21st International Conference Information Visualisation (IV)\",\"volume\":\"102 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2017 21st International Conference Information Visualisation (IV)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/iV.2017.31\",\"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 21st International Conference Information Visualisation (IV)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/iV.2017.31","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
摘要
新型高性能硬件的出现也推动了对新编程模型的需求。开放社区运行时(Open Community Runtime, OCR)建议使用基于任务的编程模型来针对其中的一些体系结构。在OCR中,整个程序从头到尾都需要使用任务来表示,并使用任务对任务的依赖来同步,这极大地限制了现有方法在应用程序开发和调试中的适用性和有用性。在本文中,我们提出了基于应用程序执行的跟踪数据来可视化任务及其同步的方法。通过这种方式,应用程序开发人员可以将预期的任务组织与OCR运行时系统所看到的实际依赖项进行比较。
Visualization of Open Community Runtime Task Graphs
The emergence of new types of high performance hardware also drives the need for new programming models. The Open Community Runtime (OCR) proposal uses a task-based programming model to target some of these architectures. In OCR, the whole program from start to end needs to be expressed using tasks and synchronized using task-to-task dependences, significantly limiting the applicability and usefulness of existing approaches to application development and debugging. In this paper, we present our approach to visualizing tasks and their synchronization, based on trace data from application execution. This way, the application developer may compare the intended organization of the tasks with the actual dependences as they are seen by the OCR runtime system.