Indico Data | A Crunchy Case Study

Indico Data Saves on Database Costs with Crunchy Postgres for Kubernetes

compared to Amazon RDS
50% cost reduction

Indico Data is a data company working with banks, insurance companies, and the healthcare industry to automate document ingestion. Indico Data uses its own LLM and AI tools to equip customers with data-centered solutions for decision making, pricing, and research. They have seen continued success in their market, processing over 10 billion items a year, and along with this growth, their data platform is increasing by 10x per year.

Reduce database costs

Indico Data was hosting their databases in Amazon RDS before switching to Crunchy Data. While they enjoyed the full-featured nature of RDS and the worry-free backups, the costs were spiraling out of control. Indico Data fully isolates each customer data environment, so they were dealing with scaling issues both on the cluster side and on the storage side. With their number of clusters growing and the size of their data growing, they began looking for a new database solution. “Our baseline costs were just too high, and going up every year,” said Don Zereski, SVP of Engineering at Indico Data.

Find a cloud native Postgres solution

Indico Data knew that they wanted their new Postgres solution to be cloud native. Their team had deep experience in Kubernetes and their DevOps team was excited about using Kubernetes to manage their clusters. They knew they’d be taking on a bit more engineering time to self manage clusters, but they suspected that this additional overhead would be worth it.

We have a terrific DevOps team and we were looking for the right tool for them to manage databases with minimal operational overhead.

Don ZereskiSVP of Engineering at Indico Data

Indico Data was also managing databases across Azure and Amazon, with customer deployments in each. They wanted a solution that worked in multiple clouds and architecture environments and after conducting research for a Postgres provider for Kubernetes, they landed immediately on Crunchy Data. They did some initial testing with sandbox clusters, benchmarking, and then started making plans to integrate the Crunchy Postgres for Kubernetes into their product suite.

We had run a few things ourselves with Postgres in Kubernetes and really liked how much easier the Operator was to handle. We didn’t even test other operators. We didn't feel like there was anything that would be as reliable as Crunchy Data.

Don ZereskiSVP of Engineering at Indico Data

Indico Data needed high availability and were excited that a primary and failover option, across multiple availability zones, would be included out of the box. They also needed reliable backups that were easy to use. They really like working with pgBackRest and have developed their own backup restore and testing process.

Crunchy Postgres for Kubernetes: It just works

Since adopting Crunchy Postgres for Kubernetes, they’ve been really happy with the outcome.

Crunchy Postgres for Kubernetes just works. We have so much else going, we don't want to think about the databases. We want the database to just work so we can focus on our other priorities. After moving to Crunchy, we don't spend time thinking about or talking about the databases.

Don ZereskiSVP of Engineering at Indico Data

Don’s team has been doing quarterly check-ins with the Crunchy Data support team but those are very quick. “We don't need a whole lot of support,” said Don. They’ve been able to manage quite a few projects in house, including a full major version upgrade. Their team has found CPK really easy to work with and integrate into their entire Kubernetes DevOps suite.

Cost Savings on Crunchy Postgres for Kubernetes

The Indico data team has been really happy with the cost savings. They knew they would take on a bit more labor costs but they estimated this at only a small percentage of a FTE. “We found that for our application, it has been worth our time to spend on the management aspects to reap the savings”, said Don.