Hacker News new | past | comments | ask | show | jobs | submit login

Noticeably missing are any details about concurrency control and replication or recovery protocols. A Twitter clone is one thing but any sort of application needing ACID Transactions is a whole other beast.



All data on Rama is replicated automatically with a configurable "replication factor". Data written to Rama is not made visible until it's successfully replicated. The documentation we're releasing next week includes a page going into detail in how this works.


Does "replicated" here mean that the data is persisted to disk?




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: