Tesora’s OpenStack DBaaS Supports MongoDB, Cassandra, Redis, and MySQL

Add Your Comments

Based on the open-source Trove project, Tesora has built a scalable OpenStack Database-as-a-Service platform that supports MongoDB, Cassandra and Redis databases in addition to MySQL databases.

The Tesora DBaaS Platform is scalable, enterprise-grade, and gives users the benefits of OpenStack Trove. It also adds features such as automated file configuration at installation, tools for deploying guest images of database instances, and free guest images from the Tesora website for MySQL, Cassandra, Redis and MongoDB.

Tesora CEO Ken Rugg said, “I think it’s really important that a project like Trove and our Database-as-a-Service platform is actually bringing the capabilities of operations and replication and visibility to all these kinds of newer database technologies.”

Scalability Shouldn’t Factor into the MySQL or NoSQL Dialemma

While its NoSQL capabilities are certainly an innovation, another major reason for using Tesora DBaaS is to provide more features to traditional MySQL databases – such horizontal scale-out.

“Why do people choose the NoSQL route?” Rugg said. “Some of it’s because they want a simpler, more basic interface that they can use to interact with it in a more dynamic schema or a document-oriented approach like MongoDB offers, but at the same time, some people are just going that way because they can’t get the scalability from the relational databases because they weren’t really designed for scale-out.”

And that’s not necessarily the right reason to choose NoSQL over MySQL, he says. “If your application is well-suited to NoSQL, that’s great. We want to be able to enable that in our Database-as-a-Service environment. But at the same time, if your application is better suited to a relational schema, and the tools that surround that, we want to make that a solution you can easily deploy as a Database-as-a-Service, but also scale out and create that kind of highly scalable tool.”

For MySQL, Tesora DBaaS has been able to scale-out TBs of data on a single deployment and reach speeds of 1 million inserts per second.

And when workloads are orchestrated over different servers, the systems have actually been found to have better-than linear scalability. “We actually see super-scalar scalability where five servers can do more than five times as much as a single server,” said Rugg. This is partly due to the fact that individual systems work better dealing with small amounts of data and memory.

Giving Back to Trove

Tesora has also been taking what is has been discovering and contributing the code to the Trove project. Companies like HP and Rackspace have also embraced Trove and pushed its code further for their own implementation. But Tesora’s Trove contributions, Rugg said, are around “making it easy for service providers or enterprises to adopt it by making it easy to install and easy to configure.”

Making Databases Easier for Organizations

A commercially-supported DBaaS offering for OpenStack allows many companies to overcome not only technical challenges, but also organizational bottlenecks.

For instance, between sending a request for a database and having it up and running could take weeks or months. And those experimenting in NoSQL might chose to use an outside service like Amazon EC2 for their MongoDB database without notifying IT. And going outside of regular channels could mean lack of oversight and potential security breaches.

Tesora does away with the burden of handling many complex administrative tasks, and allows users to quickly create relational and non-relational database databases as needed.

Add Your Comments

  • (will not be published)