{"title":"Optimizing Memory Allocation in a Serverless Architecture through Function Scheduling","authors":"Manish Pandey, Young-Woo Kwon","doi":"10.1109/CCGridW59191.2023.00056","DOIUrl":null,"url":null,"abstract":"In a serverless architecture, a function does not fully utilize the allocated memory. Such memory over-allocation increases node utilization and wastes resources, causing cold-start and latency issues. This paper presents a fine-grained scheduling approach for a serverless architecture that aims to address the issue of over-memory allocation and improve data locality. The proposed approach estimates how much memory each function uses so that similar functions can be scheduled on the same node. As a result, it makes less use of each node and keeps the state within a single node. We evaluated our approach through the existing FaaS applications and real-world data.","PeriodicalId":341115,"journal":{"name":"2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 IEEE/ACM 23rd International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CCGridW59191.2023.00056","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
In a serverless architecture, a function does not fully utilize the allocated memory. Such memory over-allocation increases node utilization and wastes resources, causing cold-start and latency issues. This paper presents a fine-grained scheduling approach for a serverless architecture that aims to address the issue of over-memory allocation and improve data locality. The proposed approach estimates how much memory each function uses so that similar functions can be scheduled on the same node. As a result, it makes less use of each node and keeps the state within a single node. We evaluated our approach through the existing FaaS applications and real-world data.