Sramana Mitra: You’ve already segued into this. Give us a sense of what you do at Scalr.
Sebastian Stadil: We have an open-source enterprise cloud management platform that we sell to customers. What we sell is actually a subscription to support – meaning a constant stream of bug fixes.
Sramana Mitra: What would be the competition for Scalr?
Sebastian Stadil: The primary competition to Scalr would be do-it-yourself [segment] where developers feel and are very much empowered to build their own cloud management platform. In many ways, they can be successful at and in many ways, they can be divesting resources away from their own customers.
Sramana Mitra: This doesn’t sound completely convincing to me. Let’s take a couple of use cases and explain precisely what you do.
Sebastian Stadil: Let’s take the example of one of our largest customers, which is Samsung. Samsung’s got close to half a billion Android devices in circulation. Those devices collect a lot of data and Samsung takes that and aggregates that to be able to run analytics on it. They need large quantities of cloud resources to be able to ingest all that data and process it. The management of all those thousands and thousands of servers requires tooling that will achieve multiple goals and enforce security practices, and also have a lot of orchestration tools. When a developer wants to run a new experiment or a new analysis over some data, they will need to set up a cluster for that analysis and the design and prototyping of that is done through Scalr. Scalr is a management platform for cloud resources and sometimes is used to be able to manage the infrastructure that’s required to perform analytics on all the data that they get.
Sramana Mitra: You help manage all the data that’s coming into Samsung’s cloud from all these devices all over the world?
Sebastian Stadil: Not quite. We help manage the cloud resources. We don’t have anything to do with smartphones or data. Let’s take it this way. If you’re using Amazon as your tool, we’re going to be provisioning instances or servers. When you have two or three servers, you can manage that in your head. You know what server does what. But when you have thousands of servers and those are being provisioned by hundreds of developers, it starts to get difficult to know which server does what and coordinating clusters of servers as well as ensuring role-based access control – meaning who can do what. It becomes a tough problem.
Sramana Mitra: What part of this do you do?
Sebastian Stadil: Scalr does the management of those resources. It’s software that allows you to know what inventory of servers you have and allows you to enforce management practices over those resources. Let’s say you’ve got a big supermarket. You’ve got lots of people that are responsible for different aisles. You might have somebody that’s responsible for the fresh fruits or toys. You are dedicating specific individuals to be able to manage the inventory of those sections. Imagine that each one of those items in that supermarket is a server or an IP address. All of those resources need to be accessed. Just like in supermarkets, you don’t want to have a team of folks that have the general mission of just doing things. You want to be able to assign specific tasks. Scalr, who manages that entire infrastructure, can then be grouped into specific people and will have specific authorizations to perform specific actions.