The new solution is said to have no theoretical limit.
Microstrategy has announced the availability of what it claims to be the largest ever in-memory analytics service.
CEO Michael Saylor used the business intelligence firm’s annual Las Vegas conference to boast that the new parallel relational in-memory engine (PRIME) has “no theoretical limit”, meaning customers could use it to manipulate hundreds of terabytes of data.
The solution is designed to enable businesses to cheaply and quickly produce Big Data-driven apps, by combining the power of both parallel and relational databases.
Saylor said: “PRIME is meant to take everybody in the analytics space to the next level. We want to break through into a new regime where we literally have terabytes of information at our fingertips in seconds.
“It’s like [previous in-memory business intelligence product] OLAP Services, but on steroids.
“We’re going to see over the course of the next few years hundreds of companies emerge with large networks of users generating extraordinary volumes of data. And they’re going to need products like PRIME to keep command of those networks and harness them.”
He suggested that the in-memory solution outperforms SAP’s HANA by running three times as fast and supporting as many as seven times the number of users.
But Microstrategy president Paul Zolfaghari later told CBR his firm is not competing with the enterprise resource planning giant, which just pushed back its operating margin target to 2017 in order to invest in becoming the leading business cloud provider.
Zolfaghari said: “It’s not designed to go head to head with them. We’re very well integrated with SAP HANA ourselves, in that we have some of our premier customers running Microstrategy against HANA from our cloud.
“So we are not really competing with database vendors. PRIME is an enabling engine in the Microstrategy layer, making our environment faster and more performant. Customers like Gucci still have a database, which they still use, even Facebook. They use Microstrategy too.”
Facebook CIO Tim Campos said PRIME was in fact developed partly as a result of Microstrategy’s work with the social network giant to analyse 30TB of raw data.
He said: “That’s 175 billion rows of data, and you can access this in seconds, the average query response time is less than three seconds on this system.”
PRIME will be available from next week in the company’s own cloud, while it should be available on the enterprise marketplace from 2015 for integration with other cloud platforms, according to Saylor.