What is Scalability in Cloud Computing

What is Scalability in Cloud Computing

Enabling customers to rely on Hadoop today, while building toward our long-term vision

We founded Pepperdata in 2012 with the vision of enabling every company to benefit from large-scale distributed computing on huge volumes of rapidly-changing data. Sean and I had already built, operated, and witnessed the power of such systems at a few of the world’s top Internet and enterprise software companies.

A decade from now, every enterprise will be running distributed systems throughout their business, and it should be just as easy to install, run, and use a massively distributed application as it is today to install and use an application on a single computer (or even a phone!). Making this vision a reality requires a distributed supervisor to control and allocate hardware usage across the distributed computing fabric in real time, and that is what we’re building at Pepperdata.

While we had this long-term vision from the start (and the vision is not limited to Hadoop, or even data systems in general), when we began building the distributed supervisor, we wanted to make sure our product was delivering real value to real customers along the way. Hadoop clearly represented the best platform to do that; by 2012, cutting-edge tech companies were already using Hadoop, and in the years since, Hadoop adoption has exploded. Pepperdata’s current product, our cluster optimizer for Hadoop, is the first version of the more general distributed supervisor we’re building, and its early success has followed the overall adoption of Hadoop.

Our customers are some of the largest companies that rely on Hadoop in production. Pepperdata’s cluster optimizer controls Hadoop in real time, providing operators with more control so that they can meet SLAs while getting more work done with the same hardware. We’ve seen great success in such verticals as media, advertising, consumer electronics, finance, manufacturing, and telecom. With the new round of funding we’re announcing today, we’ll be able to scale our sales and marketing team and accelerate our product innovation. We’re both scaling up and scaling out: we’re adding depth to our product line, and we’re also bringing the product to customers in new geographies and verticals.

Wing VC: Building companies that create new markets

As we scale the current product and bring our vision to new customers, we’re also keeping our eye on the long-term goal of making distributed computing fabrics both easy to operate and enterprise-grade. That’s one of the reasons we’re so excited about working with Peter Wagner and the rest of the Wing team. Of course Peter has a track record of investing in great enterprise companies, but more than that, he has a passion for building companies from an early stage with a focus on companies and technologies that create brand-new categories.

Wing’s focus is a perfect fit for our approach of starting with a concrete product that solves real problems for real companies while building toward a world where distributed computing at web scale is ubiquitous, reliable, and valuable for every enterprise. When we started working with Peter, we were impressed by how deeply he engaged with both the hard technical problems we’re solving and our long-term vision of how Pepperdata’s distributed supervisor will enable entirely new distributed computing applications and ecosystems by making it easy for companies to buy, deploy, and use the equivalent of today’s shrinkwrap software on tomorrow’s distributed computing fabrics. (Peter wrote a great blog post today explaining his view of where Distributed Computing Fabrics are headed, and how Pepperdata is a key enabling technology.)

Complementing Wing’s vision with boots on the ground

We’re also tremendously excited about having several new strategic investors join this round. We’ve been working with the Citi Ventures team for quite a while now, and we’re excited about forming a closer partnership. Citi is using Hadoop to create value from data