We are always on the hunt to learn about new and interesting technologies.  Case in point, back in April of 2018, VMblog had the opportunity to speak with Simon Crosby, chief technology officer at a new startup called SWIM, when the company launched its edge intelligence software company out of stealth mode, while at the same time announcing its SWIM EDX platform.  Fast forward almost a year later, and we're catching up with Crosby to find out the latest.

VMblog:  We spoke during the launch of SWIM EDX last year.  Catch us up.  How was 2018?

Simon Crosby:  Absolutely - it was a banner year for us. As you know, we launched the company and our edge intelligence software in April of last year, and in the first year alone we've grown by leaps and bounds. We also closed a Series B round of funding in July, led by Cambridge Innovation Capital with participation from Arm and Silver Creek Ventures. It's been an incredibly exciting time and with that, most of our focus has been preparing to make SWIM available to the open source community. 

VMblog:  Today you are announcing the SWIM platform is now open source.  What problems are developers facing and how does SWIM address them?

Crosby:  Developers run into several problems trying to work with real-time data, mainly because most application architectures are combinations of a message broker, an application server and a database. This causes them to spend an enormous amount of development time integrating these components and then optimizing them to work as sufficiently as possible. 

Swim is useful to developers because they can incorporate multiple different data streams relatively quickly. Developers who are building Swim applications no longer have to worry about the individual components for various software activities - like applying application logic or messaging amongst different nodes within the system. This is a much quicker development process because developers don't have to integrate all those different components into their application, which they then might have to repeat both on the edge and then in the cloud.

VMblog:  Why is it important for applications to process higher volumes of data now more so than compared to past years?

Crosby:  They will need to be able to process this high volume if they want to achieve new efficiencies. The world is constantly speeding up and new streaming data sources are being integrated into business applications, coupled with the rapidly growing demand for real-time applications. As the software world continues to bend toward real-time, stream processing has become critical to most enterprise applications. To succeed, they will have to process higher volumes of data than ever before.

VMblog:  SWIM.AI focuses on its ability to build stateful, real-time, distributed applications.  Why are these attributes important?

Crosby:  Software has been trending more towards real-time collaborative experiences over the past few years. Learning from customers and from my efforts at past companies I've worked with, real-time is becoming more and more important to application developers because it's necessary to generate insights within that time context. With its ability to build real-time applications, the Swim platform allows developers to process data sooner, and because Swim is universally stateful, it can process data locally at the source. Being stateful also means Swim can introspect itself as data flows across the system and because every Swim service is known for its application context, it can also monitor for anomalies in the software's behavior.

VMblog:  Can you talk about which industries will benefit most from SWIM?

Crosby:  Swim has already been deployed across several different industries. We have customers in the smart city space, manufacturing, asset tracking and logistics. Across all those industries, we've seen a core pattern that people are struggling with - they have high volumes of real-time data in several formats from a heterogenous number of locations or sites and they want to build an application which combines these things to be able to analyze and optimize their processes.

One example in the smart city space is the work we've done with the city of Palo Alto, California and our customer Trafficware to integrate with their connected traffic systems. The challenge they had was that the sensors at each intersection were generating high volumes of data, making it difficult to store in their central traffic management center. Swim was able to help them by processing data locally at the edge, which significantly reduced their cloud cost while also improving the reliability of their systems.

VMblog:  How can developers start using SWIM?  And what should they expect in terms of a learning curve?

Crosby:  The first step they can take is by going to developer.swim.ai where they can download Swim right away. From there, they can download the Swim library and start testing it on their local machine. The Swim community portal includes all the support developers need to start building applications. We also offer SDKs for Java and Javascript, video tutorials and several reference examples for applications we've already built.

VMblog:  Anything else we should know?  What can we expect from SWIM in the coming months?

Crosby:  Now that the Swim platform is open source, we are looking forward to experiencing a new company dynamic and working with a new community of developers. We have also expanded our product development team so we expect to be branching out into more new industries and geographies very soon.

##