Choosing The Right Data Store

Andrew Ross (LocationTech)

09:00 on Friday 20th September (in Session 22, starting at 9 a.m., EMCC: Room 1)

Show in Timetable

Description: An architectural view of application data model and data storage technologies.
Abstract:

So you're going to build the next killer app? That's awesome! Flexible hosting solutions from Google, Amazon, Rackspace, and many others make it really easy to get started. You can reuse powerful open source libraries and services to focus your efforts on developing something unique. Ideally your technology choices support your application's architecture and your business model. Ideally you can scale up or out as your needs change. There are so many good technology choices, so which one(s) do you pick? This presentation focuses on storage technologies. It is a mixed database and architecture talk. We will examine some of the popular storage and processing technologies such as PostgreSQL, MongoDB, SQLite, Hadoop, and others. We'll play match maker and try to pair up the model each technology is ideal for. Do you need a small footprint for mobile devices? Do you low latency for web or mobile applications? Do you need a high degree of redundancy? What proportion of the time does your application read vs. write data? Do you need to optimize CPU for processing large data sets? Do you need to minimize bandwidth or disk space? What about support for spatial data types and spatial functions? After attending this talk, you should have a good idea of some of the important factors in choosing your data storage technology and a rough sense of some of the great open source options available to you.