This is a guest post written by Dan Scholnick, General Partner at Trinity Ventures, an early stage venture capital firm dedicated to partnering with passionate entrepreneurs to transform. Scholnick is a General Partner at the firm and focuses on Cloud Computing, Consumerization of IT, Mobile, Enterprise and Infrastructure Software companies. Scholnick’s original post can also be found on the Trinity Ventures blog here.
Here at Trinity, we’ve been following the DevOps movement for quite some time, at least since we made our initial investment in New Relic in 2008. Despite our conviction and New Relic’s tremendous success, we’ve found that there’s still confusion as to what the fuss is all about. As a software engineer turned venture capitalist, the DevOps movement is one of the most exciting trends I’ve seen in IT in the last decade. I thought I’d share our perspective on why this is so important and how it’s guiding our investment strategy.
What’s all the excitement about?
The DevOps movement is all about the convergence of software development and IT operations, driven by cloud computing. In the traditional software lifecycle, developers wrote the code and then tossed it over the wall to an operations team to get it running in production. Ops engineers would rack servers, configure networks and install software, then make sure everything stayed running. That all changes in the world of cloud computing , where infrastructure becomes a programmable, fungible commodity. A software developer using cloud infrastructure like AWS no longer needs traditional ops. The code calls Amazon’s APIs, provisioning and configuring infrastructure as needed. New versions of the software deploy all-new instances of the supporting “hardware” and everything cuts over seamlessly. The core skill then becomes writing great code – and the operations and development happen in tandem.
This changes everything
Why does this matter to a venture capital firm like Trinity? The DevOps engineer needs a different set of products and tools than her IT ops counterpart – tools that aren’t available from the established systems management vendors. This creates huge opportunities for innovative startups, two of which I’ll highlight from our portfolio.
New Relic’s core application performance management (APM) product provides developers and DevOps alike an unprecedented level of insight into how their software is running. Unlike traditional systems management tools that help ops engineers manage on-premise infrastructure such as servers and networks, APM focuses on the health of the application wherever it is. Over a decade ago while at Wily Technology (now part of CA), Lew Cirne and I helped solve the APM problem for traditional IT organizations. Today two things have changed. First, in the world of cloud, you need an APM solution that can manage an app across a distributed and elastic infrastructure environment. Second, the application itself has become more important than ever. When infrastructure is fungible and elastic, all the DevOps engineer cares about is the health of the app – new infrastructure is just a push away. New Relic elegantly solves these problems, providing deep insight and awareness for developers, as well as DevOps engineers focused on performance and health.
In 2010 we led the seed round for Docker (formerly known as dotCloud) for one simple reason: DevOps means that the way applications are packaged, deployed, and run is fundamentally changing (though Docker’s business model has evolved since its early days as a PaaS vendor, the fundamental premise is the same). Rather than requiring custom configurations and painstaking management, Docker “containerizes” applications components such that every container is lightweight and behaves consistently. Applications and their underlying components can be programmatically deployed, managed and moved on ever-changing cloud infrastructure without a hint of operating system or hardware configuration. In a pre-Docker world, companies with tremendous and evolving application demands looked to virtualization as a way of abstracting their infrastructure, but paid a tax in dollars and performance for doing so. In the future we think of Docker will take the mantle as the VMware of the DevOps world, with containers as the ultimate DevOps platform.
That’s not all
Obviously, Trinity isn’t the only venture firm on the planet to have figured out that the DevOps movement is disrupting an established ecosystem that could be fertile ground for startups. There are a number of exciting venture-backed companies looking to make infrastructure into something programmable. Chef and Puppet automate hardware configuration using code. Circle makes testing and publishing software as easy as pressing “go.” In our portfolio, Loggly andCloudscaling are bringing the most stubborn parts of an IT organization into the cloud, making them important pieces of programmable infrastructure.
Get on the bandwagon
The combination of cloud, mobile, and agile development are changing the IT landscape in a big way. DevOps represents a new discipline, one that has its own challenges and will need its own set of tools. We’re still in the early days of this movement and there are plenty of problems left to solve. Expect to see Trinity make more investments in world-class entrepreneurs who have a compelling vision for where the market is going.