PWL Beijing : Zhang Kai on PacificA Replication and Jiang on Amazon Aurora

This is a past event

3 people went

Location image of event venue

Details

We will be talking about two classic papers of distributed storage systems in this event.

PacificA: Replication in Log-Based Distributed Storage Systems

An influential paper describing the experience in designing and implementing replication for commonly used log-based storage systems. It gives a general replication framework that is simple, practical, and strongly consistent.

This paper will be talked by Kai Zhang, Cofounder & CTO of SmartX.

https://www.microsoft.com/en-us/research/wp-content/uploads/2008/02/tr-2008-25.pdf

Amazon Aurora: Design Considerations for High Throughput Cloud-Native Relational Databases

A very famous paper by Amazon about the design of massively-deployed Amazon Aurora, cloud-native database. It describes how doing so not only reduces network traffic, but also allows for fast crash recovery, failovers to replicas without loss of data, and fault-tolerant, self-healing storage.

This paper will be talked by Jiang, backend engineer of Xtalpi.

https://www.allthingsdistributed.com/files/p1041-verbitski.pdf

1500, Tower L, Hailong Building, 1 Zhongguancun St, Beijing, China

我们这次 Papers We Love 活动,将于 8 月 11 日下午在中关村海龙大厦进行,分享计算机科学中存储和数据库领域的两篇经典论文,包括的主题是:

* PacificA: Replication in Log-Based Distributed Storage Systems (张凯,SmartX 联合创始人及CTO,可以关注他的知乎专栏)

PacificA是微软为大规模分布式存储系统开发的一个通用复制框架,该框架简单,实用,提供强一致性,并且可以适配不同的复制策略。它对于我们理解分布式系统的强一致性,构建工程化的分布式系统有很好的指导意义。

https://www.microsoft.com/en-us/research/wp-content/uploads/2008/02/tr-2008-25.pdf

* Amazon Aurora: Design Considerations for High Throughput Cloud-Native Relational Databases (江嘉诚,Xtalpi 后端工程师)

Amazon Aurora 云原生数据库服务的论文,有独特的数据复制和云原生设计,并且也是计算存储分离和 share-everything 架构的经典。

https://www.allthingsdistributed.com/files/p1041-verbitski.pdf

时间:2019 年 8 月 11 日(星期日) 下午 14:00 - 17:30

地点:北京中关村海龙大厦 L 座 15 楼[masked] Xtalpi 办公室

交通:地铁 4 号线中关村站 E 出口,海龙大厦 L 座星巴克的入口