Features

Cost

Accessing the benefits of parallel computing has, up until now, required very specialised engineering skills – the ability to effectively program FPGAs using VHDL/Verilog etc. – that are costly and hard to find.

Being able to program FPGAs with Go means you can use your existing skills and workforce to design parallel computing into your projects right away. Switching to parallel computing will also have the knock-on effect of consolidating costly hardware requirements because one FPGA can do the work of many servers. We provide pay-as-you-go access to our service so you only pay for the compute time you need.

Speed

Probably the primary reason to choose parallel computing is speed.

We all use multi-core processors, where a handful of processes can happen at the same time, one per core. But, using parallel hardware, such as an FPGA, you can have 100s of processes running simultaneously. We use Go’s in-built concurrency features to allow you to do just that, by deploying to FPGAs in the cloud. You can identify areas of your project that are well-suited to work on parallel hardware, and make that happen, with no new skills or hardware required.

Familiar Tooling

Our familiar tooling is key to us providing a great user experience.

We’re aiming to provide an interface that you feel comfortable with so you can quickly start deploying your code. We have a simple command line tool, really helpful error messages and you deploy to cloud-based FPGAs with a simple command once your Go code has built. All our code examples are held in a public Github repo too so there’s always inspiration for new projects.

Another key advantage to using FPGAs is that they are fully reprogrammable, so you can make subtle or big changes to your programs whenever you like. This leads to a really productive workflow – you can try things out and make improvements easily. We also have a hardware simulator and a quick code check command so you can get really quick feedback on your code without having to do a full build.

Programs coded in go

Quick-check locally before upload

Reconfigure.io

Simulate, build, deploy

AWS EC2 F1 instances

CPU FPGA

Compare Features

Choosing a processing method depends on the goals of your project. Click through the various options below to compare their features.

Documentation

Our documentation provides guidance on all aspects of our service, from workflow and tooling to help with troubleshooting your programs. There are also tutorials to help you get started.

Learn more

Join our mailing list

Add yourself to our mailing list to receive news and updates from Reconfigure.io.