Clavis Insight provides online and e-commerce insights and analytics to help leading consumer packaged goods companies, such as P&G and Unilever better how their products are performing in the market. After just 5 years, customer acquisition and new product development dramatically increased the company’s data ingestion rate. Clavis’s infrastructure needs inverted from 80% compute and 20% storage to 40% compute and 60% storage. "We realized we needed to pull the storage out of the cloud or it was going to break the bank," says McDaid.
The company engaged with Logicalis, a NetApp partner in Ireland, who proposed the optimum solution based on Clavis’s specific requirements. “Migrating their storage to an on-premises facility while keeping compute in AWS was the best solution for Clavis,” explains Mick Kehoe, chief technologist at Logicalis. “Leveraging NetApp Private Storage and Cloud Volumes ONTAP was a perfect fit for their needs.”
When we started testing Cloud Volumes ONTAP, we found we needed 2/3 less storage and that meant 2/3 less cost.Garreth McDaid Lead Development Operations Engineer, Clavis Insight
Before moving its data, Clavis Insight tested the solution with Cloud Volumes ONTAP in its AWS infrastructure. The savings were immediate. "We got all these benefits from Cloud Volumes ONTAP. Deduplication and compression alone reduced our storage requirements and costs by 67%," says McDaid. "It was amazing."
Another benefit came with NetApp FlexClone® thin-cloning technology. In the AWS environment, Clavis Insight development teams all used a single central database. "People were stepping on each other’s toes trying to do development projects," explains McDaid. "FlexClone technology enables us to give each team their own copy of the database without any storage impact. It’s helping us speed product development dramatically, and that translates to a competitive advantage."
Cut cloud storage costs by
Get to know what NetApp customers already know: Right now, the smart move is a digital transformation made possible by a data fabric.