As a public cloud service optimized for Big Data applications, GoGrid gives developers access to a number of open source platforms, including Hadoop, Apache HBase, Cassandra, MongoDB, and Riak. Trying to fuel what GoGrid CEO John Keagy describes as an open data services (ODSs) market, GoGrid provides the ability to stand up applications on multiple platforms in a way that helps developers avoid getting locked in to one particular architecture. Now GoGrid is close to taking that concept one step further by exposing an API to the GoGrid cloud platform.
In the process of building a new 1-Button Deploy portal, GoGrid created an API. Once GoGrid makes that portal available, customers will have the option of either using the portal or invoking its services directly via an API.
Keagy says that GoGrid is creating a polyglot cloud for Big Data applications and that GoGrid provides a level of orchestration services that make it easier to manage the cloud environment. In effect, Keagy says that GoGrid is trying to bring all the benefits of a platform-as-a-service (PaaS) environment to the development of Big Data applications. As part of that effort, GoGrid also announced today partnerships with Basho, DataStax, Hortonworks, and MongoDB, under which instances of the Big Data platforms provided by these companies will be supported on GoGrid.
Instead of standardizing on one type of Big Data platform, Keagy says it’s clear at this point that developers will be building multiple types of Big Data applications, which, based on the attributes of the application, are going to be better suited for one platform or another. Rather than having to acquire all the hardware needed to stand up those applications, GoGrid provides access to a variety of Big Data platforms without requiring customers to hire the expertise needed to manage each of those platforms themselves.
Currently serving 15,000 customers, GoGrid has no shortage of competitors. Unlike most of their competitors, GoGrid has invested in the solid-state drives (SSDs) and 10G to 40G Ethernet networks needed to drive the performance of Big Data applications, Keagy states.
When it comes to Big Data, it’s hard to say with any certainty which platforms developers are going to need access to in the future. Nevertheless, chances are that a broad range of SQL and NoSQL interfaces will end up being invoked not only at different times, but also in conjunction with each other. GoGrid clearly wants to be the one-stop shop in the cloud for making any one of those services a simple API call away.