Abstract | ||
---|---|---|
Current weak consistency semantics provide worst-case guarantees to clients. These guarantees fail to adequately describe systems that provide varying levels of consis- tency in the face of distinct failure modes, or that achieve better than worst-case guarantees during normal execu- tion. The inability to make precise statements about consistency throughout a system's execution represents a lost opportunity to clearly understand client application requirements and to optimize systems and services ap- propriately. In this position paper, we motivate the need for and introduce the concept of consistability—a uni- fied metric of consistency and availability. Consistabilit y offers a means of describing, specifying, and discussing how much consistency a usually consistent system pro- vides, and how often it does so. We describe our ini- tial results of applying consistability reasoning to a key- value store we are developing and to other recent dis- tributed systems. We also discuss the limitations of our consistability definition. |
Year | Keywords | Field |
---|---|---|
2008 | worst-case guarantee,client application requirement,distinct failure mode,keyvalue store,initial result,consistability definition,normal execution,current weak consistency semantics,consistent system,consistability reasoning,failure mode,weak consistency | Eventual consistency,Computer science,Position paper,Weak consistency,Consistency model,Semantics,Distributed computing |
DocType | Citations | PageRank |
Conference | 6 | 0.96 |
References | Authors | |
20 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Amitanand S. Aiyer | 1 | 457 | 19.60 |
Eric Anderson | 2 | 348 | 27.02 |
Xiaozhou Li | 3 | 164 | 8.55 |
Mehul A. Shah | 4 | 3547 | 317.66 |
Jay J. Wylie | 5 | 685 | 44.29 |