Architecting for Big Data: a presentation by local company MIOsoft

Hi everyone,

Here are the links for the two presentations for the last meetup. Thanks to everyone who came. Apologies for the delay in getting these out.

http://www.slideshare.net/RyanBosshart/pig-hands-on-november
http://tiny.cc/pythonhadoop

As I stated at the last meetup there is no meeting in December. Out next meeting will be Tuesday, 28 January at 6 PM in room 302 of the Central Library. Here is an abstract of the presentation. I hope to see everyone there!

Pitt

When architecting for Big Data, you have to think outside the box. We will explain how an IBM project at CUNA Mutual Group in 1998 morphed into MIOsoft and planted the seeds of innovative ideas that eventually yielded a novel database paradigm called the Context Database that further facilitated the creation of a tightly-integrated distributed application development ecosystem for Big Data. We will contend that MIOdb is capable of storing and analyzing Big Data while simultaneously being capable of scaling to meet the demands of high-throughput transactional applications because of a unique combination of features, such as:
• Lock-free operations
• Logical and physical clustering of data
• Context-oriented concurrency
• Parallel reporting requiring only a single physical pass across each disk
• First-order modeling of complex, deeply-nested structures
• Automatic management of denormalization
• Sophisticated data integration concepts like transitive matching of like data, auto-relationship discovery, and conservation of data provenance
We will close by looking to the future, the “Internet of Things” and real time contextual analytics, and detailing how a Context Database platform like MIOedge makes understanding individual entities in real time possible.