Data integration vendor Fivetran is expanding its portfolio of data source connectors with a series of new additions.
The new data source connectors became generally available on Jan. 19 and include connectors for IBM DB2 and SAP HANA as well as an improved connector for the Oracle Database.
Fivetran’s platform helps organizations more easily enable enterprise data integration for analytics and business intelligence applications.
The market for enterprise data integration is growing, which has led to investor interest in Fivetran. The vendor raised $100 million in a round of funding in June 2020.
Fivetran data connectors at Autodesk
Among the organizations that expect to use the new Fivetran data connectors is software vendor Autodesk, which is well known for its AutoCAD software platform.
Jesse Pedersen, vice president of data platforms and insights at Autodesk, explained that the vendor uses Fivetran to enable data pipelines that feed into business intelligence and analytics platform that help Autodesk run its business.
“A lot of what we’re doing is to enable the business to run more efficiently,” Pedersen said. “We made the transition from selling perpetual licenses to subscriptions a few years ago and I think in a subscription world, having correct information and appropriate information on your customers is vital to keeping them happy.”
Pedersen was previously a co-founder of construction software vendor BuildingConnected, which was acquired by Autodesk in September 2019. He noted that when he joined Autodesk, nearly all of the data was being stored in Amazon S3, yet not all of the data was being analyzed.
That’s where Fivetran comes in, enabling Autodesk to build data pipelines that bring in enterprise data for analysis. Currently Autodesk is building some 20 different data pipelines and has plans to double that number over the course of 2021, thanks in part to Fivetran’s data connectors.
From manual processes to automated enterprise data integration
Without an automated data integration provider like Fivetran, Pedersen said Autodesk would have had to use a time-intensive and manual process to integrate data. Pedersen said that when he first joined Autodesk he was told that it could take up to six months for a new data pipeline to be built, which he said wasn’t acceptable.
Jesse PedersenVice president of data platforms and insights, Autodesk
“Now the key performance indicator that we measure is from the time that a group comes to us with a data ingestion request to the time where that first bit of data lands in Snowflake, ‘can we get there in six days?'” Pedersen said.
As to why Pedersen chose Fivetran instead of a different vendor that could help with enterprise data integration, a key factor is the vendor’s singular focus. He noted that other vendors often include data integration as a part of a larger platform but all he really was looking for was a targeted product that could help with extracting, loading and transforming (ELT) data to enable a data pipeline.
“I love the fact that ELT is Fivetran’s sole focus as a company,” Pedersen said.
New connectors expand enterprise data integration
Among the new connectors is one for SAP HANA, which will be particularly useful for Autodesk, Pedersen said. Pedersen noted that Autodesk is an SAP user for ERP and the SAP HANA database is a core part of the vendor’s back office systems.
“We’re excited to see Fivetran enable the SAP connector as we’re going to be pulling in more and more financial data into our data warehouse,” Pedersen said.
While the SAP HANA and IBM DB2 connectors are entirely new, the Oracle connector is an improvement on a previous release.
The improved Oracle connector adds the Flashback transaction query mechanism for replication, explained Fraser Harris, vice president of product at Fivetran. He noted that Fivetran had earlier supported the LogMiner approach to getting data from Oracle.
“Flashback is a much more efficient native method to capture changes,” Harris said. “Our end goal is for every database source connector to behave the same — a fast copy of the table with automatic schema creation, automatic data type detection, automatic column mapping, automatic change data capture and automatic schema migration.”