• E-mail this page
• |  Print this page
• | 
• | 
• | 

Using Oracle Database In The Cloud

Two of the biggest announcements from the Oracle OpenWorld Conference in San Francisco this week were that customers could run some Oracle products within Amazon.com's AWS cloud computing environment, and that the Redwood Shores, Calif.-based software giant would be supporting other cloud environments in the future.

I was able to nail down some of the practical details of that announcement at a Wednesday OpenWorld session titled “Using Oracle Database in the Cloud." Roughly 150 attendees in the afternoon session heard Sushil Kumar, Oracle's senior director of product management, explain the whys and wherefores of hosting Oracle-based solutions in AWS and Dr. Peter Tonellato, head of Harvard Medical School's Center for Biomedical Informatics, show how his laboratory used Oracle and AWS to quickly develop some innovative genetic testing models.

Kumar said Oracle has partnered with AWS to offer products and services that would allow Oracle apps to be deployed in the AWS Cloud and allow Oracle databases to be backed up to the AWS Cloud. He also said that these offerings would most likely be extended to other Cloud platforms in the future. In addition to making Oracle 11g database, Oracle Fusion Middleware, Oracle Enterprise Manager, and Oracle Enterprise Linux software licensable and supported in the Amazon Elastic Compute Cloud (EC2) environment, Oracle is also making available a set of preinstalled, preconfigured virtual machine images for EC2 -- Amazon Machine Images (AMIs) -- to allow users to easily provision a fully functional Oracle environment in as little as 30 minutes. He further described how Oracle's new cloud computing-based data backup solution called Oracle Secure Backup Cloud Module used Amazon Simple Storage Service (Amazon S3) as a backup destination, and would enable encrypted data to be backed up to the Cloud, since it's based on the company's encrypted tape backup management technology. He said Oracle and Amazon were also in talks to offer the option of delivering physical backup tapes in case AWS experienced outages more serious than the hiccups Amazon experienced earlier this year.

In answer to an audience member's question, Kumar said scalability -- the ability to use one small server or a thousand big ones -- and the fact that AWS doesn't require a long-term usage contract should make the Oracle AWS option attractive to enterprise users doing pilot or rapid development projects outside of normal corporate IT channels.

One such rapid development project built on Oracle and AWS was described by Dr. Peter Tonellato, who heads the Laboratory for Personalized Medicine (LPM) of the Center for Biomedical Informatics at Harvard Medical School. Tonellato's lab focuses on personalized medicine -- preventive health care for individuals based on their genetic characteristics -- by creating models and simulations to assess the clinical value of new genetic tests. To overcome the difficulty of finding enough real patient data for modeling, LPM creates patient avatars -- or "virtual" patients. The lab can create different sets of avatars for different genetic tests and then replicate large numbers of them based on the characteristics of hospital populations.

Tonellato said LPM needed to find an efficient way to manipulate many avatars, sometimes as many as 100 million at a time. "In addition to being able to handle enormous amounts of data," he said, "I wanted to create a system where postdoctoral researchers could scope a genetic risk situation, determine the appropriate simulation and analysis to create the avatars, and then quickly build Web applications to run the simulations, rather than spend their time troubleshooting computer technology." Tonellato evaluated several alternatives for the lab, including dedicated servers, a new or co-located data center, and several cloud computing options before deciding to use Oracle and Amazon Web Services.

Starting from the time Tonellato opened his AWS account on May 16, 2008 ("not the first time I've given my credit card number to Amazon," he joked), LPM was able to create a development environment for its researchers' Web applications in approximately 6 weeks. On June 6, Oracle sent Tonellato the company's prerelease Linux Amazon Machine Images (AMIs) to use with his data modeling and by June 16, Tonellato's team had finished customizing the Oracle Linux AMIs. Two weeks after that, the team had its first Web application up and running. "The combination of Oracle and AWS allowed us to focus our time and energy on simulation development, rather than technology, to get results quickly," concluded Tonellato.

It's worth noting that Oracle on-demand approach to cloud computing doesn't offer pay-per-use pricing. As detailed here, Oracle pricing is based on migrating regular upfront, fully-paid Oracle software licenses to EC2. While Oracle supports deploying Oracle Database, Middleware and Enterprise Manager products on the Amazon EC2 platform, a further wrinkle in Oracle AWS support is that Amazon EC2 is a virtualized environment that uses a custom virtualization engine that isn't supported by Oracle. In addition to assuming no responsibility for data stored in Amazon S3, the company encourages Oracle AWS users to sign up for Amazon Web Services Premium Support to get technical assistance with virtualization and other issues related to Amazon's Cloud infrastructure.

For more information:

InformationWeek has written a "Guide to Cloud Computing" that details the cloud offerings and strategies of Amazon, Salesforce, Microsoft, and five other providers.

« T-Mobile Revises G1's 3G Data Limit Policy | Main | The Evolution Of Content Creation Tools »