SCADS: Scale-Independent Storage for Social Computing Applications

TL;DR

SCADS is a scalable storage system designed for social web applications, enabling declarative consistency, cost-effective scaling, and predictive performance modeling to handle large, relaxed-consistency data workloads.

Contribution

The paper introduces SCADS, a novel architecture combining declarative consistency, utility computing, and machine learning for performance prediction in social computing storage.

Findings

Supports large-scale social applications with relaxed consistency

Uses machine learning for performance and resource prediction

Enables cost-effective scaling with declarative consistency requirements

Abstract

Collaborative web applications such as Facebook, Flickr and Yelp present new challenges for storing and querying large amounts of data. As users and developers are focused more on performance than single copy consistency or the ability to perform ad-hoc queries, there exists an opportunity for a highly-scalable system tailored specifically for relaxed consistency and pre-computed queries. The Web 2.0 development model demands the ability to both rapidly deploy new features and automatically scale with the number of users. There have been many successful distributed key-value stores, but so far none provide as rich a query language as SQL. We propose a new architecture, SCADS, that allows the developer to declaratively state application specific consistency requirements, takes advantage of utility computing to provide cost effective scale-up and scale-down, and will use machine learning models to introspectively anticipate performance problems and predict the resource requirements of new queries before execution.