Hadoop may make many promises, but Teradata delivers solutions that enterprises trust.
We’re not saying this to ruffle any feathers because it isn’t an either or game. Every now and then we see headlines that say things like “Cloudera Declares End Of Data Warehousing Era” or see pictures of tombstones with RDBMS written on them, but the reality is that a good many of Teradata’s users ignore them or figure that Teradata will deliver big data capabilities, in a palatable doses, when the time is right.
And that’s essentially what Teradata and Hadoop pioneer Cloudera are banging the drums about this morning as they announce an expanded partnership around technology integration, sales and support. In other words, the two companies will work together to integrate Teradata’s integrated data warehouse and Cloudera’s enterprise data hub so that customers can work with multiple data sources (Cloudera’s Enterprise Data Hub being one of them) through the Teradata Unified Data Architecture (TDA).
This means that not only will Teradata customers be able to “buy” Cloudera’s products and services from Teradata, but that they’ll be tuned to work together and mutually supported.
A Look at the Partnership
For those who follow the Hadoop vendors closely, it’s an interesting turn of events, but before we get into that, here’s how Teradata and Cloudera customers should benefit from the relationship:
Deeper Software Integration. Teradata has created connectors to integrate with Apache Hadoop as part of its Unified Data Architecture. These connectors are now certified with Cloudera Enterprise, allowing both technologies to work together and extend mutual support for new integrations. Two key components of the UDA are:
1. Teradata QueryGrid, which lets business analysts ask any question, easily pulling together data from both platforms into a single answer set.
2. Teradata Loom, which provides tracking, exploring, cleaning, lineage, and transformation of Hadoop Distributed File System files. It helps operations managers and programmers easily evaluate and manage millions of HDFS files.
Licensing. Under the agreement, Teradata is licensed to resell Cloudera Enterprise solutions, including Cloudera Enterprise Basic, Flex and data hub editions, in addition to other Hadoop distributions.
Training, Customer Support and Consulting Services. Teradata also licenses Cloudera education and training services. Teradata Customer Support will provide maintenance and support for CDH. Teradata Professional Services for Hadoop and the Teradata data warehouse are easily deployed on Cloudera Enterprise – services such as data cleansing, governance, data lineage, predictive analytics, and security.
Haven’t We Seen This Before?
In 2012 Teradata announced a similar partnership with Hortonworks and, from what we can tell. it seems to be going well. At that point, we believed that Teradata had made a deliberate choice to integrate this deeply with a single Hadoop partner who had little to no proprietary code at all.
If that’s not the case, we can’t help but wonder why it took so long for Cloudera and Teradata to develop their relationship to this extent, except maybe comments like the ones below made by Cloudera co-founder Mike Olson, around six months ago, as he spoke to Jessica Twentyman at ComputerWeekly.
“Everyone wants to know if Cloudera is after Teradata,” he (Olson) said. “The fact of the matter is that the opportunity here is not to knock over old guys and steal their wallets. The opportunity is to monetise vast amounts of data using tools that weren’t previously available.
That doesn’t mean there won’t be a place for data warehouses, or that products from EDW suppliers won’t evolve over time, he said. “But workloads that belong in high-end enterprise data warehousing systems today, won’t in the future – and even high-performance, interactive analytic workloads will run in Hadoop.”
Could it be that the love that Teradata’s customers hold in their hearts has softened Cloudera up a bit? Or that Cloudera’s Enterprise Data Hub is so sexy that Teradata wants a ride even if it doesn’t last very long?
There are, no doubt, so many possible explanations, what’s yours? And will MapR be next?