Columnar databases are designed for high performance queries and analytics. This session will cover the differences between row and column databases, and how Infobright's columnar database, built on MySQL, delivers high performance without indexes, data partitioning or other DBA effort. It will also discuss how to migrate from traditional row-based products, and present several case studies.
This tutorial will provide an in-depth tutorial on various forms of NOSQL (NotOnlySQL) datastores (key/value, data structure store, document store and wide column stores) for working with semi- structured data. The data ranges from web logs to social and knowledge graphs to configuration data stores for cloud infrastructures and other domains.
Moore's Law has run its course, yet despite the growing demands placed
on databases, traditional solutions offer little alternative to vertical
scaling. Come learn step-by-step how to use Apache Cassandra to turn a
cluster of inexpensive commodity servers in to a massively scalable
Google App Engine is an development & hosting platform that lets you build & deploy web applications on Google's high-traffic infrastructure. You only need to upload your code: no more worrying about machines, storage, scalability! This tutorial introduces attendees to its architecture & various service APIs. In the hands-on lab, you'll build+deploy a real app to the cloud using Python in minutes!
MongoDB (from "humongous") is a high-performance, open source, schema-free document-oriented database.
Like most web applications, memcached and MySQL formed the data foundation beneath Farmville - until mid-2010. As the popularity of that application skyrocketed, a more effective system was needed to sustain FarmVille's 500,000 operations per second. In response, NorthScale, Zynga and NHN developed _membase_ - a distributed, key-value database that is 100% compatible with memcached.
A non-classified case study that describes how we've built a stack based on MALLET, Hadoop/Cassandra, and Flare/Flex to build a highly scalable system for the U.S. intelligence community: MALLET lends itself to state of the art NLP, Hadoop/Cassandra yield a massively distributed back end, and Flare/Flex provide the tools for creating a great UI/UX capable of performing advanced analysis.
The need for database systems that scale efficiently has led to many alternatives to the traditional RDBMS. This talk presents an overview of these new non-relational databases, collectively referred to as "NoSQL," followed by an in-depth examination of SourceForge.net's deployment of MongoDB, an open-source NoSQL database.
NoSQL (or NOSQL -- Not Only SQL) is sometimes justly criticized for being too broad a category, but after thirty years of the relational database being the instinctive choice for data storage, publicizing the concept that One Size Does Not Fit All is a Good Thing. This talk will present some axes along which to evaluate database products, applied to some of today's popular NoSQL products.
Google App Engine is an development & hosting platform that lets you build & deploy web applications on Google's high-traffic infrastructure. You only need to upload your code: no more worrying about machines, storage, scalability! This session introduces attendees to its architecture & various service APIs. Time-permitting we'll go through a simple example using Python.