Not long ago, if you walked into a service provider organization and asked what people thought about open-source, you’d get a lot of blank stares. You might run across engineers playing around with open-source software in their spare time. But on the list of strategic priorities, most service providers treated open-source projects as interesting lab experiments—if they thought about them at all.
Things have most definitely changed. Today, most cloud services (and most web-scale companies delivering them) run largely on open-source server software. Indeed, the fastest-growing areas of tech (Big Data, machine learning, microservices, virtualization) largely run on open-source. Despite its historically slow evolution relative to other technologies, networking itself is moving to open-source too. The world’s leading communications service providers, and even most of the world’s leading networking vendors, now participate in multiple open-source networking projects.
What changed? Why has open-source emerged as the preferred model for so many technologies—even among traditional stalwarts of proprietary solutions, like network operators? And what do companies hope to get out of it? Let’s take a closer look.
Drivers of Open-Source
To understand the value of open-source, start by looking at its opposite: closed, proprietary solutions. It’s instructive to look specifically at the world of networking, where proprietary solutions still often dominate. At the highest level, networking isn’t a complex concept—moving packets from Point A to Point B. Yet historically, each networking vendor had its own proprietary solution for how networks were architected and operated. Ask the network vendors, and they’ll say they do this to deliver superior features and functions that work basically out of the box. Ask the service providers and large enterprises buying the equipment, however, and you’ll get a different answer: proprietary solutions let vendors lock customers into their ecosystems.
If you’re a company with a smaller network that does what you want, that may not be an issue. If you’re a Tier-1 service provider spending billions annually on networking, and you’re locked into buying from one vendor because no one else can viably compete for your business, you’re liable to get frustrated.
First, whenever you want to update your network to deliver new capabilities to your customers, it takes a very long time. Your network roadmap—in many ways, your business—is in your vendors’ hands. These vendors historically deliver new releases on the timescales associated with physical hardware, on the order of months and years. Meanwhile, the cloud and application providers you’re competing with are living in a world of agile software sprints, where changes happen in days, sometimes in hours.
More importantly, you’re stuck using networking products that, by design, are not interoperable. When you’re buying tens of thousands of network elements, you’d prefer to have vendors competing for your business, both to keep prices low and to keep your vendors responsive. But if you’re using Vendor A for your metro network, and you’d like to try bringing in Vendor B, it’s incredibly difficult. So your capital and operational costs are always higher than they would be if there was true competition.
The open-source model, especially as applied to the network, empowers companies to break out of the boxes that proprietary vendors have historically kept them. They can begin to create a networking paradigm that’s much more flexible and interoperable, and ultimately, more cost-effective. Next week’s blog will dive into tapping into the value of open source.
Learn more about how Lumina Networks applies open source solutions to return network control to service providers here.