/images/avatar.png

一直是阵雨🌦️

目前该博客不再更新,最新内容已请在新博客 zhengyua.cn/new_blog 中查看,暂不考虑迁移兼容

MIT6824LEC3笔记:GFS

GFS 课程地址 Before gfs why is distributed storage hard High performance -> shard data over many servers Many servers -> constant faults Fault tolerance -> replication replication -> potential inconsistencies Better consistency -> low performance it can sum up the all thinking points like: performance fault tolerance replication consistency these points have connection and influence each other. what would we like for consistency? There are two conditions

MIT6824LEC4笔记:Primary-Backup Replication

Primary/Backup Replication 课程地址 the note from the lecture Primary/Backup Replication for Fault Tolerance Case study of VMware FT, an extreme version of the idea fault tolerance to provide availability despite server and network failures using replication What kinds of failures can replication deal with? “fail-stop” failure of a single replica fan stops working, CPU overheats and shuts itself down someone trips over replica’s power cord or network