The Aerospike Vision
Srini Srinivasan, Chief Development Officer and Founder, Aerospike, discusses the Aerospike vision.
Aerospike was founded to solve a really important internet problem. We decided there was a better way to build a database that can handle internet-scale applications. And that’s essentially our founding principle, is to take advantage of the latest in hardware technology to enable very high-performance database applications.
Flash immediately allowed random access to storage at scale. What this meant is, with the right kind of database re-architecture, especially in terms of the file system, we could now build a database which could work across tens of terabytes of data, while still providing low milliseconds in terms of read and write rates, and that’s really critical. A read-write database is extremely important to implement internet-scale applications, and then we could build one which was 10 times faster and 10 times lower in terms of TCO at scale.
The problems that Aerospike solves are big problems, scale problems. And they exist in mature enterprises which are already profitable and growing, and then they hit a wall. And the wall is what Aerospike helps climb for these enterprises.
One of the most important characteristics of a database is its ability to keep data safe. Aerospike has demonstrated in production its industry-leading performance already for almost six years of uptime. Now, with the addition of strong consistency, Aerospike provides linearizability, which is the strongest form of consistency, without actually reducing significantly the performance which was available already. And thus, such a database which can combine such high performance with very high consistency has not been available in the marketplace before, which means the applications that allow using multiple layers in order to achieve these results can collapse into one Aerospike installation. And this, in turn, will make these applications simpler to build, easier to upgrade, and also more profitable to deploy.
Aerospike’s future is going to be, in my opinion, being a technology that companies can depend on for a decade or two in the future. What that means is, we are essentially committed to solving the hardest problems inside enterprises in terms of mission-critical real-time access to data, which is also needed for better customer experience in virtually every case that we have seen so far.
So we see Aerospike as a technology that is going to be the fundamental building block for such highly consistent applications which enable a new class of user experience which has not actually been very prevalent so far.