#1 => TAO: Facebook's Distributed Data Store for the Social Graph

This time it will be very informal, we will meet up at the coffee house Stofan Café (https://www.facebook.com/stofan.cafe) as they have comfortable couches, beers and good coffee.

Instead of having a speaker giving a talk on the paper we will simply try to explain it to each other and have some fun discussions that almost always go way off topic :)
Reading the paper is highly recommended but don't stress it.

--
The paper we will discuss is "TAO: Facebook's Distributed Data Store for the Social Graph".

"A single Facebook page may aggregate and filter hundreds of items from the social graph. We present each user with content tailored to them, and we filter every item with privacy checks that take into account the current viewer. This extreme customization makes it infeasible to perform most aggregation and filtering when content is created; instead we resolve data dependencies and check privacy each time the content is viewed. As much as possible we pull the social graph, rather than pushing it. This implementation strategy places extreme read demands on the graph data store; it must be efficient, highly available, and scale to high query rates."

• Nathan Bronson

Abstract:

We introduce a simple data model and API tailored for serving the social graph, and TAO, an implementation of this model. TAO is a geographically distributed data store that provides efficient and timely access to the so- cial graph for Facebook’s demanding workload using a fixed set of queries. It is deployed at Facebook, replac- ing memcache for many data types that fit its model. The system runs on thousands of machines, is widely dis- tributed, and provides access to many petabytes of data. TAO can process a billion reads and millions of writes each second.

Paper link: https://facebook.com/download/273893712748848/atc13-bronson.pdf