DataStax pivots back to its original strategy
The DataStax and Cassandra stories are somewhat confusing. Unfortunately, DataStax chose to clarify them in what has turned out to be a crazy news week. I’m going to use this post just to report on the status of the DataStax product line, without going into any analysis beyond that.
Pro tip: If you choose to announce at a conference where many other vendors will surely announce news also, you naturally run the risk of not garnering much attention.
For starters, it may help to realize or recall that:
- Cassandra was originally developed and revealed at Facebook, to much early NoSQL fanfare. Facebook later backed away from Cassandra use.
- Rackspace guys in Texas became Cassandra’s biggest backers. They eventually founded a company called Riptano to commercialize Cassandra.
- Texas company Riptano became the California company DataStax.
- DataStax came out with a Hadoop-on-Cassandra offering called Brisk. For a while, it sounded as if Hadoop was as big a focus for DataStax as Cassandra is.
- DataStax is now recommitted to being the Cassandra company, and has accordingly backed away from Hadoop and Brisk as a separate or coequal focus. However, it sees Hadoop capability as a nice, or even major, feature of its Cassandra-centric offering.
- To finalize its open source obligations with respect to Brisk, DataStax is in essence:
- Donating a Hive driver for Cassandra straight into the main Apache Cassandra project.
- Releasing the rest of Brisk as a separate open source project.
- Disclaiming interest in further advancing open source Brisk.
- There’s also something called Solandra — evidently SOLR-on-Cassandra — whose status is similar to Brisk’s.
- There are three main ways that DataStax helps you to consume Cassandra.
- DataStax is the principal sponsor of Apache Cassandra development, and presumably long will be. Apache Cassandra is both free-like-speech and free-like-beer.
- DataStax is also introducing a paid-subscription version of Cassandra called DataStax Enterprise, which features proprietary code, support, and so on. DataStax Enterprise is neither free-like-speech nor free-like-beer.
- There will also be something called DataStax Community Edition. DataStax Community Edition is free-like-beer, but not free-like-speech.
Various posts on the DataStax blog give DataStax’s explanation of what it’s doing. Ben Werther, the ex-Greenplum guy who briefly worked at DataStax and was most associated with telling the Hadoop/Brisk story, has moved on to his own startup Platfora.
DataStax Enterprise has three main aspects:
- DataStax Server, which is the actual database and analytics code. At this time, there is little closed-source code in DataStax Server, but DataStax reserves the right to widen that gap in the future.
- DataStax OpsCenter, which is management tools around DataStax Server. DataStax OpsCenter is entirely closed-source, even though DataStax gives a limited version away for free.
- Support.
To describe DataStax Community Edition, I’ll just quote the press release verbatim, which characterizes it as:
… a free platform based on Apache Cassandra that bundles the open source database with smart installers, drivers and connectors for popular development languages, demo apps, documentation, and a free version of DataStax OpsCenter for Apache Cassandra.
DataStax Community Edition is crippleware only in terms of feature set; there are no limitations on its database size, cluster size, or usage rights. A core mission of DataStax Community Edition is to create happy Cassandra users, who may then become customers for DataStax Enterprise.
Comments
5 Responses to “DataStax pivots back to its original strategy”
Leave a Reply
What you have written is completely incorrect. DataStax is not backing away from Hadoop at all. Brisk is not dead, but is reborn in a larger suite of pay-for products. It’s pivoting to invest more in Brisk and make it a larger focus. They just received $11 million in series B funding and much of that is focused on investing in what Brisk is becoming – DataStax Enterprise. Please research and ask the company before writing something so inaccurate.
Jeremy,
I stand by my analysis, which is based on extensive conversation with DataStax’s top management.
But then, what you’re criticizing is something different than what I actually wrote.
What I got from the article was that DataStax was backing away from Hadoop. That is not true.
I stand by the sentence that used the phrase “backed away”, just as I stand by the rest of the article.
I don’t necessarily stand by your misreading of that sentence.
I apologize. I just know that there is already some misunderstanding in the community and the tweet that brought me to this post reflected that. I’ve just tried to work hard to help the Cassandra/Hadoop integration along for the past year and a half and didn’t want there to be misunderstanding on that front after a lot of work.
Thanks for checking me on the rest of the article.