Rafay enables developers to automate the distribution, operations, cross-region scaling and lifecycle management of containerized microservices across public/private clouds and service provider networks. Rafay’s platform is built around foundational elements that together deliver an optimal abstraction layer across disparate infrastructure, making it easy for developers to scale and operate their apps across any number of locations (regions).
So why does Rafay’s platform need to exist?
Because running apps across a number of regions is really hard! I wrote a blog in Forbes where I discussed the obvious and not-so-obvious challenges with deploying apps across regions.
A few super-successful companies such as Netflix and Uber have internally built (and now maintain on an ongoing basis) custom platforms that help them scale their apps across any number of regions across a mix of provider infrastructure. The investments in people and time required to build out and operate such complex platforms are significant. Netflix and Uber can do this, but can you?
Adrien Joly at Algolia wrote an excellent blog recently describing the pain he and his colleagues experienced in moving a simple web crawler from Heroku to a managed kubernetes service (GKE) on GCP. Adrien and team are very capable engineers who, given time, can solve any infrastructure problem. But if running a relatively simple containerized workload can be this hard across a couple of clusters sitting next to each other, consider the pain to run a more complex scenario where apps are being run across multiple regions, possibly in multiple infrastructure provider environments. Someone in your company now has to manage a multitude of (k8s) clusters on an ongoing basis. I don’t mean just bringing up the clusters, but deploying apps, securing the environment, getting secrets and configs in place as a service, solving for upgrade strategies across regions, solving for log aggregation across disparate networks, and so on. Over time, you’ll end up with a stack that looks like the following:
Whether you are deploying a simple web crawler across multiple regions for reliability like the team at Algolia, or running a cloud gaming platform across multiple public clouds, or operating a ride-share platform, the level of specialization needed to build and maintain a unified application distribution, cross-site scaling, and life-cycle management has net-negative ROI.
Even if you have built a number of tools and processes in-house already, consider that the battle-hardened engineers who maintain your platform (your platform will evolve over time because the underlying software will also evolve) could be leveraged to work on strategic initiatives instead of spending time re-inventing the wheel. Or if you are expanding your footprint beyond a single public cloud provider and plan to use additional cloud providers’ infrastructure, Rafay can augment your team’s in-house tool set with an easy-to-use platform that you can bring up in any cloud environment today.
This is why Rafay’s platform exists. The platform lets you deploy containerized apps anywhere, and helps you achieve reliability, availability and performance through a developer-friendly interface. Be it AWS, GCP, Azure, Equinix, or a service provider network, you can now run your containerized microservices in the environment that makes most sense given your business’ needs. Every feature you will end up building or cobbling together over time to run such a platform in house, we provide as a service.
And here’s the key issue every line of business owner should be considering: You use public clouds instead of deploying your own servers in a nearby colocation facility (colo). Why? Because the ongoing investment in people, time and tooling required to maintain colo environments is impossible to recover for all but a tiny fraction of companies. So when it comes to building and maintaining multi-site app distribution and operations platforms, which direction makes most sense?