DataStax launches Constellation, new cloud platform for Cassandra - here’s why...

Profile picture for user ddpreez By Derek du Preez May 21, 2019
Summary:
NoSQL database provider DataStax is entering a new era, where it recognises that Cassandra needs to be built with multi-cloud and developers front of mind.

Image of Datastax logo

The NoSQL database market appears to be reaching a level of maturity aimed at the mainstream cloud buyer, that perhaps wasn’t prevalent just a couple of years ago. The latest announcement from DataStax, at its inaugural user event in Washington DC this week, moves the company from one that is purely focused on the flexibility and capabilities of the open source Apache Cassandra database - and instead considers what may be necessary within an enterprise where there is a large developer team working in a DevOps environment, and one that needs ease of use, implementing applications with multi-cloud in mind.

The announcement this week being DataStax Constellation, a multi-cloud platform built to offer Apache Cassandra and services around the database. Cassandra has become popular amongst businesses that are deploying complex applications that are built for the internet, due it its distributed nature and being ‘always available’. We see more and more customers moving away from traditional relational models to NoSQL database companies, such as DataStax.

I got the chance to speak with Billy Bosworth, CEO at DataStax, about the company’s latest announcement, which clearly marks a shift in direction that takes into account the following:

  1. Developers like to use NoSQL, open source databases, but they want them to come with easy-to-use management tools (this has driven a large part of MongoDB’s success, a key competitor to DataStax)

  2. Multi-cloud is increasingly becoming the deployment model of choice for the enterprise. Companies are happy to shift between Azure, AWS and Google Cloud Platform, depending on the use case, and the data layer needs to accommodate for that.

The announcement

DataStax claims that Constellation is the only cloud platform specifically architected for Apache Cassandra - which it will now offer as-a-Service, the core component of the platform - allowing “easy development and deployment of Cassandra-based applications while reducing the overhead from database operations in the cloud” via:

  • A single, integrated web console with capabilities such as single sign-on for all available cloud services

  • DataStax Studio, an interactive developer tool for CQL (Cassandra Query Language), which aims to make it easier for developers to collaborate by keeping track of code, query results and visualisations in self-documenting notebooks.

Speaking with CEO Billy Bosworth, he said that the announcement is the beginning of a “multi-year vision”, which will provide the foundation for future development.

Focusing on improving the operational simplicity for developers, he added:

We have taken on some of the most difficult and challenging data workloads on the planet. And that has led us to having a reputation for being able to solve most distributed data challenges out there in a very elegant way. And so, power and scale are two things readily associated with DataStax and Cassandra. And that's good. We think that that's a real testament to our architecture and a real testament to the power of the engineering engine behind the scenes

But what that does is open us up to a whole new world of people that won’t have that power and don't want to have to go to the depth of understanding on how to do the operational management behind that architecture. And so, this is new for us in the sense that making things easy and simple, which hasn’t always been synonymous with Cassandra.

This opens us up to a whole new base of developer. They might have been attracted to our power in the past, but put off by the operational architecture. Now we will be able to court and engage them in a very simple way.

Insights and multi-cloud

The second component of the Constellation platform (thus far) is DataStax Insights, which is centred around performance management and monitoring for DataStax Cassandra and DataStax Enterprise in the future. It will include centralised monitoring across all cloud and on-premise deployments, an at-a-glance “health index”, AI-powered analysis and recommendations that enable automated performance tuning, and cloud-through discovery to simplify troubleshooting.

However, apart from having Cassandra as-a-Service, the other most appealing aspect of Constellation is its multi-cloud commitment. Although only available on Azure and Google Cloud Platform at the moment, it is thought that Constellation will soon be available on AWS too.

Bosworth explains:

The next piece that is really important is that all of this works because of the desire for hybrid, and multi cloud environments and our customer base. That's why this makes sense for us. What they want is the ease of use on the front end for the development teams for sure, but also that enterprise customers are going to be working out their hybrid and multi-cloud architectures for years to come. Having a consistent data layer with the same interface that gives you that compatibility and application portability is very important to them.

It's increasingly clear to me that multi cloud is how customers are seeing their future. I wouldn’t have said that two years ago. Two years ago there was still a lot of questions about what workloads were going to be moved to the cloud, let alone settling on more than one cloud.

And that's non-trivial for our customers. So we need to make that very easy when it comes to the data because the data tends to be the hardest part to get right in this environment. It's not it's not a trivial problem - trying to figure out how to make your data available across different cloud infrastructures. We need to make that easy.

My take

This announcement aligns very closely with two key trends I’m seeing at the moment from buyers. Firstly, multi-cloud is the status-quo. That is also the case for those operating purely in the public cloud. Whilst some workloads are best suited for AWS, others are best suited for Azure, and customers want to have the choice to place their workloads where they see fit [read: where’s cost effective]. Secondly, developers will stick with you if you can make their lives’ easier. That’s one of the primary reasons MongoDB has been so successful. Whether or not DataStax can execute on both these points concurrently, whilst also innovating on the platform core, remains to be seen - the likes of Google Cloud Platform are making similar moves. However, if it can, the future investment decisions for existing and potential customers just got a whole lot easier.