{"title":"Design and Research of the Whole Process Non-blocking Technology in High Concurrency Scenario","authors":"Xue-hua Liao, Chunqi Luo, Peng Lu, Pengcheng Xiao, Zhousen Zhu, Meng Lei","doi":"10.1109/ICCCS57501.2023.10150980","DOIUrl":null,"url":null,"abstract":"Under high concurrency scenarios, software applications can suffer from low access efficiency and high exception rate due to I/O bottlenecks and database connection number limitations. Without changing the hardware conditions, a fullprocess non-blocking technology is employed to alleviate the performance bottleneck of software applications under high concurrency scenarios from the perspective of optimizing access request reception and data read/write rates. Firstly, to mitigate the huge concurrency pressure on database caused by a large number of non-essential operations, a Bloom filter is selected to filter invalid access requests. Then, a concurrent data processing caching strategy is applied to hand with the performance bottleneck problem caused by the limitation of database connections in the high-concurrency scenario. Furthermore, to address the issue that synchronous I/O cannot take advantage of the responsive framework and combine with the new achievements in database technology development, an asynchronous non-blocking database access middleware is designed and developed to achieve a completely asynchronous non-blocking and streaming data processing mode. After testing and application practice, the effectiveness and efficiency of the proposed methods are confirmed, and it also has certain wide applicability.","PeriodicalId":266168,"journal":{"name":"2023 8th International Conference on Computer and Communication Systems (ICCCS)","volume":"170 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 8th International Conference on Computer and Communication Systems (ICCCS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCCS57501.2023.10150980","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
Under high concurrency scenarios, software applications can suffer from low access efficiency and high exception rate due to I/O bottlenecks and database connection number limitations. Without changing the hardware conditions, a fullprocess non-blocking technology is employed to alleviate the performance bottleneck of software applications under high concurrency scenarios from the perspective of optimizing access request reception and data read/write rates. Firstly, to mitigate the huge concurrency pressure on database caused by a large number of non-essential operations, a Bloom filter is selected to filter invalid access requests. Then, a concurrent data processing caching strategy is applied to hand with the performance bottleneck problem caused by the limitation of database connections in the high-concurrency scenario. Furthermore, to address the issue that synchronous I/O cannot take advantage of the responsive framework and combine with the new achievements in database technology development, an asynchronous non-blocking database access middleware is designed and developed to achieve a completely asynchronous non-blocking and streaming data processing mode. After testing and application practice, the effectiveness and efficiency of the proposed methods are confirmed, and it also has certain wide applicability.