Vertica Puts Column-Oriented Database In Virtual Machine
The software runs on commodity hardware and becomes a plug-and-play addition to a data center being built out around cloud principles.
Vertica, the startup that offers a column-oriented database system for data warehouse analytics, is now offering a version of its system that runs in a VMware virtual machine.
"This is a new way to take advantage of data stored in a private cloud," said Dave Menninger, VP of product management for the Vertica Virtualized Analytic Database. It runs on commodity hardware and becomes a plug-and-play addition to a data center being built out around cloud principles, he said in an interview. If lots of commodity hardware is available, the Virtualized Analytic Database can expand its use of CPUs and solve a knotty data warehouse problem without needing a dedicated large server, he said.
"People do things in the cloud that they never would have thought of doing before. You can do things inside the corporation that you've been doing in the cloud," said Jerry Held, executive chairman of Vertica and former senior VP of the Oracle database server unit.
Vertica is competing with database machines, a database system preloaded on a piece of hardware, by offering a column-oriented database appliance, a copy of its system optimized to run with CentOS Linux in the VMware virtual machine. (CentOS repackages a version of Red Hat Enterprise Linux 5.3 or earlier.) The analytic database is designed to be used in organizations that frequently want to deploy a data mart, have highly seasonal data leading to periods of intense usage, or have analytic databases that grow rapidly, said Menninger.
It's priced at $100,000 for a 1-TB entry-level model. Three terabytes of data would lead to a higher price, but the price per terabyte declines as the number of terabytes increases, Menninger said.
Data warehouses are typically designed to run on a large server or sets of servers because of their occasional use of many CPU cycles. With a virtualized system, additional hardware can be activated as needed and loaded with virtual machines.
"Faster, more economical virtualized analytic databases will outmode specialized, hardware-based appliances," Menninger predicted. With pricing based on the amount of data being analyzed, not CPUs on which it's running, customers can achieve high performance at a lower cost.
The basic Virtualized Analytic Database stands the usual relational database paradigm on its head. It grabs information out of the columns of a table, instead of its rows. Columnar information is all similar, such as the total sales for each member of the sales force for a given month, and can be retrieved and analyzed quickly. Row information, on the other hand, may include many pieces of information about each salesperson, with the data recorded at each step dissimilar from the one that went before it.
Located in Billerica, Mass., Vertica is used by 60 customers, including companies in telecommunications and financial services. The firm brought out its first version of the column-oriented database in September 2007.
Vertica was co-founded by former University of California at Berkeley computer science professor Michael Stonebraker, who says compressing information in columns adds a performance benefit when it comes to analyzing column information. The compression is possible because one compression scheme can be applied to all the elements of the column, unlike a row.
Believe it or not, virtualization thrives in smaller companies. InformationWeek has published an independent analysis of this topic. Download the report here (registration required).
Google in the Enterprise SurveyThere's no doubt Google has made headway into businesses: Just 28 percent discourage or ban use of its productivity products, and 69 percent cite Google Apps' good or excellent mobility. But progress could still stall: 59 percent of nonusers distrust the security of Google's cloud. Its data privacy is an open question, and 37 percent worry about integration.
CIOs Get Smart About BIIT’s tried for years to simplify business intelligence efforts. Have visual analysis tools and Hadoop and NoSQL databases helped? Respondents to our 2014 InformationWeek Analytics, Business Intelligence, and Information Management Survey have a mixed outlook.