“Enterprises recognize Apache Cassandra as the de facto standard for databases at scale”
© Shutterstock / docstockmedia
DataStax is most well known for building the vast majority of Apache Cassandra. We talked with Andrew Lampitt, Senior Director, Product Marketing at DataStax about the explosion of data, the benefits of a hybrid cloud strategy and the future of Apache Cassandra and DataStax Enterprise.
JAXenter: How is DataStax balancing the needs of the community and its commercial needs?
Andrew Lampitt: DataStax contributes to open source in two major ways. First, we’re most well known for building the vast majority of Apache Cassandra, which is relied on by household names, like Netflix and Apple. If you were to sort the upcoming Cassandra 4.0 release by contributions, we’d be at the top of the list again. And we continue to build and evolve the Apache Cassandra drivers. The most important drivers are all built and maintained by DataStax as a service to the community.
Fewer people know that through our 2015 acquisition of Aurelius, DataStax also created more than 90% of Apache TinkerPop, the de facto standard graph database framework that all other leading graph databases depend upon: Microsoft, IBM, Amazon, and so on all depend upon TinkerPop. Aurelius did this in the service of their graph database Titan, which was the precursor of DataStax Graph.
Forward-thinking enterprises consider hybrid cloud as something to design in from the start.
After our founding in 2010, our efforts initially focused on making Apache Cassandra and its community both viable and thriving. We succeeded: today, Apache Cassandra is the de facto standard open source database technology at scale. But as an open source project, Apache Cassandra is very technical and complex. It requires experts to build with it and to manage its operations.
Our customers demand simplicity and ease-of-use — both for development and for operations. And they require a unified environment of database, search, analytics, and graph. They tell us that they don’t want to wire together disparate projects and web services for data management. And they need rock-solid security. So that is where we put the thrust of our efforts today, focusing on making enterprise customers successful.
JAXenter: What are the main developments in DSE6 for developers?
Andrew Lampitt: First and foremost, DataStax Enterprise 6 (DSE 6) is ridiculously fast. DSE was already the world’s leader in read-write performance, throughput, latency — everything at scale and in a distributed manner. No cloud database service or traditional relational database can come close. And we just doubled the performance for throughput and halved the latency. That translates into not just cost of ownership, but also the time you spend optimizing performance to meet your SLA in advance of going live.
Next up, developers want to build with containers. So, now DSE Images are available to pull from Docker Hub. And we provide comprehensive documentation for configuration and tips and tricks to be successful. We empower developers to work they way they want.
Now we offer DSE Bulk Loader, for loading and unloading data into and out of DSE. Remember, developers are using massive data sets. It’s a bit of a task to manually load data into Apache Cassandra. They want to get to work right away, and DSE makes it four times faster to handle that than with community tools.
Looking to the future, it’s of course cliche to speak of the explosion of data. But it’s true when you look at how the digital world is growing with tech like IoT, machine learning and AI, and how we are all now global citizens.
And then DSE Studio, our visual development environment for DSE has added a ton of ease-of-use features around collaboration (such as import/export of notebooks), SparkSQL support, and more.
Finally, this one is more on the operations side, but that matters to developers too, especially in today’s world of co-mingled devops. Remember that DSE is already amazingly simple to operate for data management at scale. You want to manage a deployment for millions of users on a global basis with 0% downtime using a relational database? First, you will need an army of DBAs and second, they will fail because sharding, replication and all those efforts to try to achieve high availability fail in the face of catastrophe. But with DSE, even a small team can manage those requirements.
What’s new in DSE 6 is self-driving operational simplicity improvements with NodeSync, TrafficControl, and an upgrade service. Now even novice DBAs and DevOps professionals can manage DSE like seasoned professionals. And that means developers can stay productive and focused on creating business value without missing a beat.
JAXenter: How are enterprises looking at using Cassandra and DSE in the future?
Andrew Lampitt: Enterprises recognize Cassandra as the de facto standard for databases at scale. If you are working at scale in a distributed fashion, you are using Cassandra. If not, you will see downtime as we saw last week with a famous gaming company that chose another technology. 12 hours down. It’s killing their business. And it’s the second time in just a couple months! I am not sure how they are coping. And then look at Sony on the other hand with PS4. If you or your family uses PS4, you’re using DSE. It never goes down.
So to answer your question about the DSE side, customers want ironclad technology, breadth of platform, ease-of-use, and a proven company with wall-to-wall experts on a global basis. That’s DataStax. They know that if Sony, McDonald’s, and even Microsoft rely on DSE, then they can too.
Looking to the future, it’s of course cliche to speak of the explosion of data. But it’s true when you look at how the digital world is growing with tech like IoT, machine learning and AI, and how we are all now global citizens. So what that all means is that developers have acceleration requirements to serve up applications at scale that demand ‘right-now’ responsiveness, no matter where their millions of users are around the world. And they are requiring that capability across hybrid cloud and every public cloud. DataStax is the only vendor that delivers on those requirements.
JAXenter: How do you think developers currently look at hybrid cloud – are they building their apps with hybrid in mind, or are they looking at internal and public cloud as separate?
Andrew Lampitt: You’ve nailed the ‘ready, fire, aim’ concern that enterprise architects and business leaders are starting to realize faces them when they look at building apps for on-premises and public cloud as separate.
There are a few things that can lead to this situation. First, they have digital transformation initiatives where they think, “massive deadline, let’s build on a public cloud database service.” But then not only is the database itself locked into that cloud, but the wiring together of services like analytics, search, and graph becomes unwieldy to manage. Then they realize, “wait, what if we need to move clouds because of partner requirements, technical advantages, or even if the cloud vendor itself becomes a competitor.”
Second, they still need to satisfy on-premises requirements, typically for cost, control, or security reasons. Now they have a cloud database and one or more on-premises databases and they don’t really talk together too well. It’s a real mess.
Consider the alternative offered by DataStax. Our customers have their bases covered. First, because they can have quick wins with new applications where DSE is a ‘data layer’ that talks to the mainframe or legacy RDBMS, but developers can work agilely with DSE for the new applications. And guess what, DSE is designed for the cloud. So that on-premise requirement can easily straddle a public cloud to hybrid cloud or even multiple clouds Second, DSE gives you what we refer to as ‘data autonomy:’ that is, both data portability as well as granular control of data for compliance, governance, and other security requirements.
So, forward-thinking enterprises consider hybrid cloud as something to design in from the start. We see it as our mission at DataStax to make that job easier.