CIOL: Why has Terracotta opened a development center in India? Talent?

AP: We were very impressed with a couple of people in India who wanted to contribute code to Terracotta and we eventually ended up offering a job to one of them. Soon after that we started seeing more and more very talented Java developers who were interested in working for Terracotta and that prompted us to start an office in India.

AP: Most people that we hire in India don't have much experience in writing OSS code. However, they have good software and Java expertise, and that is what is very important to us. We are very happy with the talent level of our Indian office. The people we are hiring in India are every bit as good as the people we hire in the US or Europe.

CIOL: How does Terracotta's infrastructure software solve a fundamental problem in the market; scaling mission-critical Java applications, while reducing the need for expensive database licenses?

AP: Terracotta allows developers to make finer grained decisions about how to share data than current-state approaches. Without Terracotta, many applications O/R-map or otherwise store Java objects in a relational database. Removing the database dependency in such cases is one way Terracotta simplifies application architecture and development.

The image depicts, there is database server and an application cluster with three servers. The database should be the size of the dashed black cylinder but it is not only storing System of Record data (labeled SoR in the diagram).

It is also working to co-ordinate between application servers. The application's scratch data is causing the database to get too big for our needs, and costs begin to escalate.

The red arrows depict the database swelling due to an overload of scratch data.

CIOL: Who are your major competitors?

AP: Our biggest competition comes from the database vendors such as Oracle. We offload the database so end up displacing database revenues.

We are also often compared with Oracle's Coherence software (formerly Tangosol), which is a distributed cache. In most use cases, Terracotta is much faster and almost always much easier to implement.