Some investors love backing ideas that could be the next big thing. Consider, for example, the rise of NoSQL databases such as MongoDB, which some market as something that scales easier than traditional relational (SQL) databases.
But what if a database existed that could support popular SQL querying but also could scale like a NoSQL database? A former engineer at Visa has built and open-sourced a database with these qualities in mind, called InfiniSQL. He’s now looking for people who’d like to test the technology, developers who want to work on it, and investors who want to bet on it.
Mark Travis started writing code for the database in his spare time about two years ago. At Visa, Travis and his colleagues used a bunch of relational databases to process transactions, including systems from Oracle and Sybase. But those systems “just weren’t up to the task, so I decided to write one myself,” Travis told VentureBeat in a phone interview.
Editor’s note: Our upcoming DataBeat/Data Science Summit, Dec. 4-Dec. 5 in Redwood City, will focus on the most compelling opportunities for businesses in the area of big data analytics and data science. There are just a few seats left, so be sure to register today!
For now the database writes information in memory, not on disk drives. What that means is access to data is quite fast, and the database can span across the memory sitting inside a bunch of servers.
On the down side, if the power goes out in a data center where InfiniSQL is running, data could be lost. He’s developing a few solutions to that problem, including writing to both disk and memory. For now, as Travis acknowledged in a post on the blog High Scalability yesterday, InfiniSQL is “still in early stages of development” and not yet ready for production workloads.
But once InfiniSQL does have means of ensuring that data is durable, companies might want to try it, just as some developers have pushed for the adoption of NoSQL databases in order to store more and more data using multiple servers instead of just one. And Travis, who lives in Santa Clara, Calif., and left Visa to work full-time on InfiniSQL in March, is already talking with friends at different companies to see about getting tests started.
If nothing else, he can point to stable performance even with more than 500,000 transactions per second and more than 100,000 connections to the database.
Travis doesn’t think his database could eliminate the need for NoSQL database. He thinks they could be useful for high-volume analytics workloads. “But operational workloads, like collecting data in real time, doing rather simple analytics, doing transaction processing, I believe that InfiniSQL — at least the design I have for it — can fit those types of workloads, really, for a large number of users,” he said.
Travis pictures a few scenarios for turning InfiniSQL into a business, including providing yearly support for the database and making it available for companies to use as a service running in a cloud.
For now, it’s probably wise to keep an eye on the project and see how Travis builds the database, who gets involved with the project, and who starts using it. That’s when investors might see an interesting opportunity.
VB's research team is studying web-personalization... Chime in here, and we’ll share the results.