Amazon Claims New Aurora DB Engine Screams With Speed

Thursday Nov 13th 2014 by Chris Preimesberger
Amazon Claims New Aurora DB Engine Screams With Speed

NEWS ANALYSIS: Aurora may become a formidable competitor for comparable database engines by MySQL, SAP, Oracle, Teradata and EMC.

Amazon Web Services, which celebrated a notable milestone earlier this year when it blew past the 1 million mark in its roster of business- and government-related customers, introduced a major new product Nov. 12 on Day 1 of its re:Invent 2014 conference in Las Vegas: the Aurora database analytics engine.

Looks like the long-popular, open-source MySQL database, which runs inside so many IT systems now that there isn't a good way to know exactly how many instances are out there, has a viable new competitor.

It also appears that super-fast databases for solid-state disks made by Oracle (TimesTen) and SAP (HANA) are also being targeted by the world's No. 1 Web services provider.

Aurora, which AWS Senior Vice President Andy Jassy described in the opening keynote as a "commercial-grade database engine at open-source cost," is a cloud service dubbed MySQL-compatible, fault-tolerant, scalable and secure with encryption.

Speed Claim: Five Times Faster Than MySQL

"The kicker is that this is generally about five times faster than MySQL databases," Jassy said. To further illustrate his point, Jassy claimed the following speed stats to back up his case:

--Aurora can process 6 million inserts per minute and 30 million selects per minute;
--it's five times faster than the largest MySQL instance;
--five times faster than MySQL on Amazon EC2, using 12 instances with eight local solid-state drives; and
--it's twice as fast as a random-access memory (RAM) disk.

"Aurora's availability is as good or better than commercial databases or high-end SANs, superior scalability and security, and we're making it available at one-tenth the cost of high-end commercial database offerings," Jassy said.

Generally, pricing per terabyte of new-gen, SSD-borne databases of this type runs likes this: $200,000 (SAP HANA); $66,000 (Oracle Exadata); $66,000 (Teradata) and $30,000 (EMC Greenplum).

Automatically Replicates Data

Amazon Aurora automatically replicates data across multiple Availability Zones and continuously backs up data to Amazon Simple Storage Service (S3), which is designed for "11 nines" percent durability, Jassy said. The database engine automatically detects and recovers from most database failures in less than 60 seconds, without crash recovery or the need to rebuild database caches, he said.

Amazon Aurora continually monitors instance health and if there is a failure, it will automatically failover to a read replica without loss of data, Jassy said.

For more details on Amazon Aurora, go here.

IT nowadays is all about speed; just ask Oracle CEO Larry Ellison, for one insider viewpoint. The increasing volume and velocity of business data and the Internet of things (IoT) require it. Storage capacity issues, for the most part, have been solved; prices for both hardware and cloudware are in tailspins. Network bandwidth pipes continue to get bigger.

Storage networking is the one area that still needs some big improvements, and if you ask Cisco Systems, Brocade and Juniper Networks, those are on the way.

Amazon is facing increasing competition just about every day in the cloud services business from the usual suspects (Google, Microsoft, IBM, HP, Dell, Oracle) and from a spate of newbie specialized companies, which have everything to gain--and generally only venture capital money to lose--at this point.

In its financials, Amazon doesn't break out revenue from AWS, but estimates from reputable sources indicates that the figure amounts to about $4 billion per year.

Originally published on eWeek.
Mobile Site | Full Site