Droplet Computing: Redefining Application Deployment

I recently had the chance to attend a briefing on Droplet Computing – courtesy of the great Michelle Laverick and Jane Rimmer (and Michelle’s most amazing cat, much to my joy).

Droplet Computing was presenting at one of the previous Cloud Field Day events (Cloud Field Day 3) and a lot of folks from our community seemed excited about it, so my curiosity was definitely piqued.

Before going on, make sure to follow Michelle for all things technical related to Droplet Computing (lots of resources, and Michelle made the jump from CFD delegate to Chief Technologist at Droplet Computing, so it must have been a truly interesting presentation there).

What is Droplet Computing

According to their website and the presentations I had, Droplet Computing is an innovative way to deploy containerized applications to multiple device platforms running major operating systems (Windows, Linux & MacOS), even if the host operating system does not supports the application.

To clarify further, Droplet Computing product allows the packaging of one or more applications into an isolated container (the droplet) that comprises of the application(s), the operating system environment on which the application(s) run, and the necessary binaries to help the droplet execute properly on the host operating system.

This means zero-touch deployment of applications, i.e. no altering of the host operating system. Running a Windows application on a Mac or on Linux becomes possible without having to install emulators or runtime environments (for example WINE on Linux).

Droplet Computing also helps running legacy applications that can run only on end-of-life / insecure operating system platforms without interacting with the host operating system.

Droplets can also execute either online or offline, but data within the droplet remains isolated from the host on top of which it executes. Users can also configure how much resources a droplet can consume by choosing the amount of CPU cores & memory allocated. During our demo, Droplet Computing demonstrated that resource requirements were quite modest to run general-purpose containerized apps.

Use Cases: Not only Legacy

Although legacy apps are a major use case for Droplet Computing, it’s perfectly possible to use Droplet Computing to deploy modern applications as well.

Droplet Computing is a great use case for BYOD, allowing organizations to deliver a pre-configured droplet with all the software needed while running independently of the host operating system, and helping achieve cost savings on hardware fleet for users.

One of the topics coming to mind when looking at Droplet Computing is VDI. A critical prerequisite to embark on the VDI journey is a thorough assessment of application performance and functionality requirements.

Those aspects touch topics such as latency and bandwidth to deliver a smooth usage experience, but storage I/O and compute needs (CPU and RAM requirements) are also critical. All of this translates into usually very large upfront CAPEX to deliver the VDI infrastructure, without a commitment that the environment will deliver.

While VDI can make sense, organizations should assess their requirements early on and weigh costs vs. benefits to decide if the VDI route isn’t overkill when all they need is just to deploy a specific set of applications to a subset of their workforce.

This technology can also be a great alternative to VDI for road warriors who often find themselves limited with inadequate connectivity.

Droplet Computing have a page listing how their solution can help by verticals & by use case.

TECHunplugged’s Opinion

Clearly there’s something exciting when looking at Droplet Computing. The technology addresses a different segment than we are used to cover, since it goes more into the end user computing space, but still we believe there’s a lot of value in the solution.

Its simplicity can be unsettling – after all, it does just one thing. But it does it well, and the creative mind can see additional use cases that are not immediately thought about.

Startups usually focus on their customers most urgent needs (that’s where the money is after all), so while there may be many ideas floating around, we can only guess about what will come next: is Droplet Computing supporting 3D acceleration? Can it be used for intensive graphical workloads? Does it support CAD applications? Can we use Droplet Computing to package games and run them on other platforms?

Of course, Droplet Computing focus is the Enterprise IT world. But I for one (as an amateur of old games, or rather someone who is often stuck with emulators or games NOT running natively on my operating system of choice) definitely see the value.

So while we wish Droplet Computing the very best in their journey with Enterprise IT customers, wouldn’t it be fun to see them contacted by gaming companies to make fully portable games across platforms? I’d gladly ditch my MacBook Pro to get a decent laptop running Linux, and run my favorites games there, courtesy of Droplet Computing. Oh, and all of my favorite collection of obsolete -yet much loved- OSes.