{"title":"简短公告:排队还是优先排队?分布式事务存储器的缓存一致性协议设计","authors":"Bo Zhang, B. Ravindran","doi":"10.1145/1835698.1835716","DOIUrl":null,"url":null,"abstract":"In distributed transactional memory (TM) systems, both the management and consistency of a distributed transactional object are ensured by a cache-coherence protocol. We formalize two classes of cache-coherence protocols: distributed queuing cache-coherence (DQC) protocols and distributed priority queuing cache-coherence (DPQC) protocols, both of which can be implemented based on a given distributed queuing protocol. We analyze the two classes of protocols for a set of dynamically generated transactions and compare their time complexities against that of an optimal offline clairvoyant algorithm. We show that a DQC protocol is O(Nlog Dδ)-competitive and a DPQC protocol is O(log Dδ)-competitive for a set of N transactions, where Dδ is the normalized maximum communication latency provided by the underlying distributed queuing protocol.","PeriodicalId":447863,"journal":{"name":"Proceedings of the 29th ACM SIGACT-SIGOPS symposium on Principles of distributed computing","volume":"30 10 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2010-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":"{\"title\":\"Brief announcement: queuing or priority queuing? on the design of cache-coherence protocols for distributed transactional memory\",\"authors\":\"Bo Zhang, B. Ravindran\",\"doi\":\"10.1145/1835698.1835716\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In distributed transactional memory (TM) systems, both the management and consistency of a distributed transactional object are ensured by a cache-coherence protocol. We formalize two classes of cache-coherence protocols: distributed queuing cache-coherence (DQC) protocols and distributed priority queuing cache-coherence (DPQC) protocols, both of which can be implemented based on a given distributed queuing protocol. We analyze the two classes of protocols for a set of dynamically generated transactions and compare their time complexities against that of an optimal offline clairvoyant algorithm. We show that a DQC protocol is O(Nlog Dδ)-competitive and a DPQC protocol is O(log Dδ)-competitive for a set of N transactions, where Dδ is the normalized maximum communication latency provided by the underlying distributed queuing protocol.\",\"PeriodicalId\":447863,\"journal\":{\"name\":\"Proceedings of the 29th ACM SIGACT-SIGOPS symposium on Principles of distributed computing\",\"volume\":\"30 10 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2010-07-25\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"4\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 29th ACM SIGACT-SIGOPS symposium on Principles of distributed computing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/1835698.1835716\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 29th ACM SIGACT-SIGOPS symposium on Principles of distributed computing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/1835698.1835716","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Brief announcement: queuing or priority queuing? on the design of cache-coherence protocols for distributed transactional memory
In distributed transactional memory (TM) systems, both the management and consistency of a distributed transactional object are ensured by a cache-coherence protocol. We formalize two classes of cache-coherence protocols: distributed queuing cache-coherence (DQC) protocols and distributed priority queuing cache-coherence (DPQC) protocols, both of which can be implemented based on a given distributed queuing protocol. We analyze the two classes of protocols for a set of dynamically generated transactions and compare their time complexities against that of an optimal offline clairvoyant algorithm. We show that a DQC protocol is O(Nlog Dδ)-competitive and a DPQC protocol is O(log Dδ)-competitive for a set of N transactions, where Dδ is the normalized maximum communication latency provided by the underlying distributed queuing protocol.