Teradata virtual storage management approach blends "hot" and "cold" storage without the redundancy introduced by Oracle's cache appliance.
Teradata has been working on fast data access for years. With last week's release of the Teradata Active Enterprise Data Warehouse (EDW) Platform 6690, the company says it delivers state-of-the-art query performance and a better approach than that offered by rival Oracle's new Exalytics appliance.
The vendor's advances in data-access speed in recent years are tied to Teradata Virtual Storage, software that monitors which data is being queried most often and then automatically moves that data to the fastest storage medium available. Before solid state disk (SSD) drives became affordable, Teradata Virtual Storage moved "cold" (infrequently accessed) data onto the inner tracks of conventional hard drive disks (HDD) and "hot" (frequently accessed) data onto the outer tracks, where faster rotation delivered quicker data access.
Teradata still uses the inner-track/outer-track technique, and it also supports high-density HDDs suitable for archival storage--super cold (very infrequently accessed) data that you nonetheless want accessible online. At the hot end of the storage spectrum, Teradata added super-fast-access SSDs back in 2010. These drives are as much as 18 times faster in data-access speeds than conventional spinning disks.
With the 6690, Teradata says there's a wider range of SSD-to-HDD configurations so customers can better tune the platform to their needs. Firms with few fast queries can dial it down to 6% of total capacity on SSDs, while firms with many such queries can crank it up to 25%. Teradata says its latest Virtual Storage software is also that much smarter, with better algorithms for learning what data to store where, with options now ranging from high-density HDDs, to the inner tracks of standard or high-speed drives, to outside tracks, to SSDs.
"The system automatically does the data placement, and it operates at the data-block level, not at the [database] table level, so it provides very granular control," said Scott Gnau, president of Teradata Labs, in an interview with InformationWeek. That granular control makes it possible to place 100% of the data needed for timely queries into SSD storage.
Teradata's chief rival, Oracle, late last month introduced Exalytics, an appliance aimed at delivering sub-second response times for data-intensive business intelligence (BI) and performance management applications.
Exalytics "adaptive caching" capabilities sound similar to Teradata Virtual Storage management in that the software monitors workloads generated by Oracle Business Intelligence Enterprise Edition-powered dashboards, queries, and analytic applications and automatically moves the hot data from Exadata (or a third-part source) into the memory of the Exalytics appliance. But there's a crucial difference, according to Gnau.
"With Exalytics it's all cache, so it's an incremental copy of data," Gnau said, describing the box as a "bolt-on Bandaid" that presents incremental storage, heating, and cooling costs. Teradata, in contrast, stores data once in the most appropriate storage option required, so Oracle is "solving a performance problem that we don't have," Gnau said.
A final 6690 platform upgrade worth mentioning is a move entirely away from 3.5-inch HDDs to smaller 2.5-inch drives. The footprint of each rack remains the same, but the 6690 can pack up to 360 drives (counting 2.5-inch SSDs) into each box. That means it offers higher storage density, lower power consumption per terabyte, and reduced cooling requirements for the total data warehousing environment as compared to Teradata's older 6680 platform.
"It doesn't sound sexy, but data-center space, power consumption, and cooling requirements are always among the top-five concerns when we survey our customers, so it's a big deal," Gnau said.
The Agile ArchiveWhen it comes to managing data, donít look at backup and archiving systems as burdens and cost centers. A well-designed archive can enhance data protection and restores, ease search and e-discovery efforts, and save money by intelligently moving data from expensive primary storage systems.
2014 Analytics, BI, and Information Management SurveyITís tried for years to simplify data analytics and 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.
InformationWeek Must Reads Oct. 21, 2014InformationWeek's new Must Reads is a compendium of our best recent coverage of digital strategy. Learn why you should learn to embrace DevOps, how to avoid roadblocks for digital projects, what the five steps to API management are, and more.