Feature Story

By Kat Rollin

July 8, 2008 — In today's fast moving business environment, you need a business intelligence (BI) solution that can help you compete with the ability to quickly analyze your data. If you already have a data warehouse solution, it might be running out of steam. If you're looking to implement a new data warehouse, you want to keep costs down. How do you replace your current BI environment or build a new one that can cost-effectively keep up with constantly changing business requirements, increasing numbers of users, ad hoc analytics, and ballooning volumes of data? Infobright can help by leveraging a MySQL database to create a simple, fast, flexible data warehouse with low cost of ownership.

The Challenge Facing Traditional Data Warehouses

The faster you can analyze data to make decisions and optimize the overall efficiency of your business, the faster you can impact revenue and increase profits. However, the amount of information is greater than ever and increases daily at phenomenal rates. Traditional data warehouses, designed to run planned reports or standard queries, are struggling under growing the demand for on-the-fly, ad hoc, complex queries with data coming from multiple data sources. These data warehouses require extensive design and ongoing management — including the need to partition and index data, which explodes 10 TB of real data into 20 to 50 TB, and requires IT to build new indices to support new queries. The usual solution is to throw more hardware at the problem, escalating the cost of the entire BI solution including IT and datacenter resources..

Radical New Approach: Intelligent Design with a Smaller Storage Footprint

Infobright overcomes the limitations of traditional data warehouses by fundamentally reinventing data warehouse architecture. Infobright's Brighthouse analytic data warehouse software achieves breakthrough performance, while eliminating the need for indices, data partitions, and other physical data structures. Brighthouse uses knowledge about the data itself to intelligently isolate the relevant information and quickly return results. Brighthouse is a column-oriented data store with an average compression rate of 10:1 and up to 40:1, which means that the size of the database is significantly smaller than the volume of raw data, and requires much less storage. Data is stored in 65K Data Packs and statistics and metadata related to the data packs form a Knowledge Grid, which is created and managed automatically by the system. Brighthouse uses the Knowledge Grid to decompress only the Data Packs needed to satisfy a given query in the fastest possible time. With this radical new approach to data warehousing, Brighthouse can answer 90 percent of queries, including highly complex and ad hoc queries, without accessing data — enabling the solution to scale up to 30 TB of compressed data using a single server

Highly Evolved and Integrated with MySQL Database

Brighthouse leverages the MySQL database, enabling you to use the mature connectors, tools, and resources associated with MySQL, and continue to harness your MySQL expertise. Because Brighthouse is integrated with MySQL, virtually all BI tools are supported including Cognos, Business Objects, JasperSoft, MicroStrategy, Pentaho, and SAS, making it easier and faster to integrate into existing BI environments. For example, Business Objects BI can connect to Brighthouse using the JDBC connector.

Infobright's Brighthouse, integrated with MySQL database, can help you dramatically lower the cost of data warehousing while enabling users to quickly query data with the speed and flexibility of Internet searches.

• Reduce storage footprint by up to 90 percent — with up to 40:1 compression
• Reduce TCO by as much as 90 percent — runs on low-cost, off-the-shelf hardware, uses fewer servers and less storage, and significantly reduces ongoing management effort and cost
• Increase response times for complex, ad hoc queries — enables user to ask any question at any time
• Reduce administration — easy to install and manage, no complex data partitioning or indexing
• Load data at very fast rates — up to 280 GB/hour
• Easily integrate with existing BI environments — with familiar MySQL tools and connectors

Infobright and MySQL

Whether you're a large enterprise trying to reduce costs and keep up with demand or a small company implementing a data warehouse for the first time, Infobright and MySQL can help you implement low-cost, fast data warehousing, so you can concentrate on creating a competitive edge.

Formerly a systems engineer for Sun, Kat Rollin now reports on partner solutions for sun.com and develops technical marketing communications for the technology industry.