Jonathan Gray, founder and CEO of Continuuity (not to be confused with Continuity Software), wanted to discuss the state of development tools used to create big data applications and what his company is doing to make the process simpler, faster and — in the company's words — "more enjoyable."
Being a reformed software engineer myself, I'm always interested in talking with someone who is trying to make the process of software development enjoyable.
Here is a summary of our conversation.
Who is Continuuity?
Continuuity is the developer of Reactor and Loom, both designed to help developers of big data applications link together a large portfolio of tools to create applications that will enable their organization to develop new understandings based upon their business and operational data; provision clusters of servers to run those applications; and then manage those clusters regardless of whether they are built in a cloud provider's data center, the organization's on-premise cloud, or on physical servers.
Gray pointed out that big data development usually involves many moving parts, such as Apache HBASE, Cascading, Apache PIG, and number of tools to support data flow and analysis. He thinks that getting all of this software to fly in formation and work together is just too hard. The company is on a mission to make big data application development accessible, powerful, fast and enjoyable.
Continuuity, by the way, is not to be confused with Continuity Software, which focuses on availability and disaster recovery solutions.
At this time, Continuuity offers two products:
- Continuuity Reactor is a "big data application server" that sits on top of the ever-growing portfolio of big data tools and makes it possible for developers to use a uniform set of APIs for application development. Reactor manages the access of the underlying tools so developers can use today's best tools and later move on to different tools without having to change their application.
- Continuuity Loom is what the company describes as "modern cluster management" that provisions, manages, and scales clusters on public clouds, private clouds, or bare metal. The company also states that Loom can deploy any application in minutes.
The discussion moved on to what the company is trying to do. In Gray's words, "developers want to play with the new and the shiny."
Continuuity has developed an application server making it possible for a developer to select tools that are "new and shiny" now and later to port them to other tools. He mentioned Apache Spark as a tool that is just emerging and is very likely to become more important to developers over time.
Another important distinction, Gray mentioned, was that Continuuity is focused on the development of real-time and user-focused applications rather than batch-oriented, back-end processes. The goal is helping organizations make big data applications useful to a broader group of analysts and decision makers.