In early 2008/late 2007, the cloud computing buzz touched the database in earnest.
More than a year later, much of that buzz has died down, as companies continue weighing security, scalability and compliance concerns about pushing their data into the cloud. The market for cloud databases is still in its early stage, but Forrester Research analyst Noel Yuhanna said there could be an upswing in adoption in the next few years.
“Today, most enterprises are struggling to reduce cost and improve manageability of their databases, and cloud databases offer a great opportunity to address such issues,” he said. “However, the technology is still maturing and is likely to take two to three years before we actually see large mission-critical databases on a hosted cloud.”
That doesn’t mean businesses aren’t taking notice. Forrester Research estimates that right now, 18 percent of enterprises are looking at cloud databases. There are a few models gaining traction today, said Burton Group analyst Marcus Collins. One is established databases such as Oracle 11g and IBM DB2 running on Amazon EC2; another is cloud databases such as SimpleDB and Google App Engine that are loosely linked to the no-SQL movement.
“Microsoft Azure adds perhaps a third model where we’ll see more traditional (i.e., relational, SQL) processing using a cloud platform,” Collins said. “This is still a work in progress as we have seen a change of emphasis of Microsoft over the past 12 months away from the prior model (no-SQL) to a more traditional model.”
Much of the activity involving traditional databases in the cloud is in the form development and testing, not running operational applications, said Matt Aslett, an analyst with The 451 Group.
“At this stage we see more interest in deploying simple applications using SimpleDB, but that is an entirely different use case from an enterprise database,” he said. “SQL Azure is interesting in that it started out as essentially a replica of SimpleDB but has become a cloud-based operational database that will enable users to extend existing and new SQL Server applications to Azure. We believe that makes sense given Microsoft’s customer base and (that) this mixed private/public cloud paradigm is likely to encourage wider deployment of database in the cloud given the launch of beta testing for Amazon’s Virtual Private Cloud, which will enable users to connect their existing infrastructure to isolated Amazon cloud compute resources and use internal security products… to protect systems running on Amazon.”
The main concerns around cloud databases – security, compliance and scalability – are essentially the same as last year, analysts said, and are part of a wider discussion about cloud computing. There are also concerns about performance issues as well – whether a particular offering can deliver sub-second query responses, for example.
“Right now vendors are not really addressing the scale issue, but mainly to establish credibility and reliability of the solution, scale will follow over the next two to three years,” Yuhanna said. “The target is to get departmental and smaller business apps on the cloud first. For example, Microsoft is positioning the SQL Azure database for small to medium sized businesses, web 2.0 and ASP. NET developers with a pay-as-you-grow database solution… We are likely to see smaller and departmental Apps in starting out over the next year, then moderate sized but less mission-critical Apps (in about one to two years), and then mission-critical App in about three to four years.”
US finalises $4.7bn award to Samsung Electronics, $1.6bn to Texas Instruments to boost domestic chip…
OpenAI begins safety testing of new model o3 that uses 'reasoning' process to ensure reliability…
US Commerce Department reportedly adding China's Sophgo to trade blacklist after TSMC-manufactured part found in…
Amazon staff in seven cities across US go on strike after company fails to negotiate,…
Two US senators ask president Joe Biden to delay TikTok ban by 90 days after…
Reporters Without Borders calls on Apple to remove AI notification summaries feature after it generates…