基于向量时钟的有界并发时间戳系统

The Bloom Clock

The bloom clock is a space-efficient, probabilistic data structure designed to determine the partial order of events in highly distributed systems. The bloom clock, like the vector clock, can autonomously detect causality violations by comparing its logical timestamps. Unlike the vector clock, the space complexity of the bloom clock does not depend on the number of nodes in a system. Instead it depends on a set of chosen parameters that determine its confidence interval, i.e. false positive rate. To reduce the space complexity from which the vector clock suffers, the bloom clock uses a 'moving…

The Bloom Clock for Causality Testing

Testing for causality between events in distributed executions is a fundamental problem. Vector clocks solve this problem but do not scale well. The probabilistic Bloom clock can determine causality between events with lower space, time, and message-space overhead than vector clock; however, predictions suffer from false positives. We give the protocol for the Bloom clock based on Counting Bloom filters and study its properties including the probabilities of a positive outcome and a false positive. We show the results of extensive experiments to determine how these above probabilities vary as …

Tiga：利用同步时钟加速地理分布式事务 [技术报告]

Tiga: Accelerating Geo-Distributed Transactions with Synchronized Clocks [Technical Report]

本研究提出面向地理复制事务数据库的Tiga设计，通过同步时钟主动分配未来时间戳合并并发控制与共识，多数场景1次广域往返即可提交事务，性能显著优于现有方案且已开源。

用于并发执行中因果序的树时钟数据结构

A Tree Clock Data Structure for Causal Orderings in Concurrent Executions

本文提出了树时钟这一新数据结构，替代向量时钟以高效计算并发程序执行中的因果序，有效克服了线程数较大时的计算瓶颈，并在 HB、SHB 和 MAZ 偏序计算中实现了显著的性能提升。

An Optimal Vector Clock Algorithm for Multithreaded Systems

Tracking causality (or happened-before relation) between events is useful for many applications such as debugging and recovery from failures. Consider a concurrent system with $n$ threads and $m$ objects. For such systems, either a vector clock of size $n$ is used with one component per thread or a vector clock of size $m$ is used with one component per object. A natural question is whether one can use a vector clock of size strictly less than the minimum of $m$ and $n$ to timestamp events. We give an algorithm in this paper that uses a hybrid of thread and object components. Our algorithm is …

点式版本向量：乐观复制的逻辑时钟

Dotted Version Vectors: Logical Clocks for Optimistic Replication

在云计算环境中，大量用户访问存储在高可用性存储系统中的数据。为了向地理上分散的用户提供良好的性能，并且即使在存在故障或网络分区的情况下也允许操作，这些系统通常依赖于仅保证最终一致性的乐观复制解决方案。在这种情况下，能够准确有效地识别并发执行的更新是非常重要的。在本文中，首先，我们回顾，并暴露问题，目前的方法，在乐观的复制因果关系跟踪：这些要么失去有关因果关系的信息或不缩放，因为它们需要副本维护的信息，随着客户端或更新的数量线性增长。然后，我们提出了一种新的解决方案，完全捕捉因果关系，同时非常简洁，因为它维护的信息，仅与注册更新给定数据元素的服务器数量呈线性增长，受复制程度的限制。