Aleksandra Knezevic, Quynh Nguyen, Jason A. Tran, Pradipta Ghosh, Pranav Sakulkar, B. Krishnamachari, M. Annavaram
{"title":"CIRCE - a runtime scheduler for DAG-based dispersed computing: demo","authors":"Aleksandra Knezevic, Quynh Nguyen, Jason A. Tran, Pradipta Ghosh, Pranav Sakulkar, B. Krishnamachari, M. Annavaram","doi":"10.1145/3132211.3132451","DOIUrl":null,"url":null,"abstract":"CIRCE (Centralized Runtime sChedulEr) is a runtime scheduling software tool for dispersed computing. It can deploy pipelined computations described in the form of a Directed Acyclic Graph (DAG) on multiple geographically dispersed compute nodes at the edge and in the cloud. A key innovation in this scheduler compared to prior work is the incorporation of a run-time network profiler which accounts for the network performance among nodes when scheduling. This demo will show an implementation of CIRCE deployed on a testbed of tens of nodes, from both an edge computing testbed and a geographically distributed cloud, with real-time evaluation of the task processing performance of different scheduling algorithms.","PeriodicalId":389022,"journal":{"name":"Proceedings of the Second ACM/IEEE Symposium on Edge Computing","volume":"52 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"10","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the Second ACM/IEEE Symposium on Edge Computing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3132211.3132451","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 10
Abstract
CIRCE (Centralized Runtime sChedulEr) is a runtime scheduling software tool for dispersed computing. It can deploy pipelined computations described in the form of a Directed Acyclic Graph (DAG) on multiple geographically dispersed compute nodes at the edge and in the cloud. A key innovation in this scheduler compared to prior work is the incorporation of a run-time network profiler which accounts for the network performance among nodes when scheduling. This demo will show an implementation of CIRCE deployed on a testbed of tens of nodes, from both an edge computing testbed and a geographically distributed cloud, with real-time evaluation of the task processing performance of different scheduling algorithms.