Dropping ACID: Architecting with Eventual Consistency in the Cloud
Share this Session:
  Jason Bloomberg   Jason Bloomberg


Wednesday, August 22, 2012
11:30 AM - 12:15 PM

Level:  Technical - Introductory

It's impossible for any distributed computer system to simultaneously guarantee immediate consistency, availability, and partition tolerance. Most enterprise DBMSs are inherently partition intolerant, thus allowing for ACID transactionality. However, Cloud Computing environments must be partition tolerant in order to achieve the elasticity benefit so important to the Cloud. Highly available Cloud applications must therefore forego immediate data consistency, and with it, ACID. What's left? BASE: Basic availability, soft state, and eventual consistency.

This session will explore how eventual data consistency affects architecting -- and rearchitecting -- applications for a Cloud environment. We'll discard the common misconception that the Cloud is a "virtual server in the sky," and explore how architects must think differently about architecting for Cloud environments vs. traditional on-premise computing.

Attendees of this session will:

  • Get a solid grounding in the CAP theorem and BASE
  • Understand the special data requirements Cloud Computing presents when architecting applications
  • Think differently about basic concepts like transactionality and consistency

Jason Bloomberg is the leading industry analyst and expert on achieving agile digital transformation by architecting business agility in the enterprise. He writes for Forbes, Wired, and his biweekly newsletter, the Cortex. As president of Intellyx, he advises business executives on their digital transformation initiatives, trains architecture teams on Agile Architecture, and helps technology vendors communicate their agility stories. His latest book is The Agile Architecture Revolution (Wiley, 2013).

Close Window