One of the main things that can be credited with Fusion-io’s success in the flash market is the company’s use of both innovative hardware and accompanying software that optimizes the offering.

A demonstration at the recent “DEMO Enterprise: An Evening of Innovation” gathering is the latest proof of the huge plus side of this approach.  Eight Fusion-powered servers managed to produce 1 billion I/O operations per second thanks to a preview of the new Auto Commit Memory (ACM) software architecture.  Specifically, the demo used eight HP ProLiant DL370s transferring 64 byte data packets, each equipped with eight ioDrive2 Duos – Fusion-io’s second generation flash storage acceleration product.

“This breakthrough is not something that could be achieved with hardware alone. Intelligent software that optimizes NAND flash as a low latency, high-capacity, non-volatile memory solution for enterprise servers can transform the way organizations process the immense amounts of data that powers our lives today,” said  David Flynn, Fusion-io Chairman and CEO.

The Auto Commit Memory ioMemory extension lowers latency to enable more performance, in addition to achieving the cost savings that come with the increased level of efficiency.

Before ACM came Atomic Writes, another subsystem extension revealed in October and covered by our own Kit Dotson. It enables what Fusion refers to as “atomic multiblock native writes,” which solves the need for two different processes to be executed in order to complete the job. This has several advantages, mainly added data integrity assurance.

Fusion-io’s fleshing out its offerings as it’s prepping up for the next 12 months. In a recent prediction published on the company’s blog it highlighted three trends that will impact the flash industry, and which it’s obviously looking to – and has already begun – to tap. Big data, or in this case small data, naturally took its spot as one of these trends.