Technology -> Infrastructure
By: David Norris, Practice Leader - Analytics, Bloor Research
Published: 3rd December 2012
Copyright Bloor Research © 2012
Big Data is one of the hot topics of the moment, and one of the most exciting aspects of Big Data is the opportunities that it offers for expanding the whole concept of Operational Intelligence. By Operational Intelligence, what I mean is that as we now have the means to design and attach sensors, which can operate in real time, into all manner of operational systems, be they electronic or mechanical, be they as large as an oil rig, as geographically spread as a communications network, or as commonplace as the family car; we can now generate and capture vast amounts of operational data.
But, as we all know, it is one thing to have data, it is another thing entirely to turn that data into meaningful insight, information that we can use to manage and improve operations. Operational Intelligence is about using that data, to not only inform us of what has happened in the past, and what is occurring at present, but, most significantly, to predict what is a likely outcome in the future from what we see now. This means that we will be able to predict and pre-empt parts failure, avoid mistakes being repeated, identify fraud before it has been completed, and maintain systems at more effective levels for longer than was previously thought possible. This is an emerging field, which today is dominated by bespoke systems being created by Systems Integrators. In Quartet FS this market is now joined by a technology that has the potential to provide a commercial off the shelf core to all such solutions, with all the advantages that brings in terms, of price, performance, resilience, future proofing, and lack of narrow proprietary lock-in.
Quartet FS has its origins in powerful analytics being applied in real time to streaming data in the Financial Services market. Typical applications have been in such things as derivatives trading, where the requirement is for highly accurate results, based not on samples, but on the whole data set. In this demanding and complex world Quartet FS has already proven itself successful, and is an established player in the global market, with a presence in London, Paris, New York and Singapore.
Quartet FS provides a parallel query capability, working in memory, using the multi-threaded capability of commodity multi-core technology. So what is being offered is a high end esoteric capability running on low end commodity hardware. About as exotic as it gets is a suggestion that, for the most demanding applications, solid state data storage is preferable to platters. This is always a winning formula, and the added bonus is that, through a partnership with TIBCO, Quartet FS can offer the TIBCO Silver Fabric technology to offer the capability across not just a single box, but a cluster, and, of course, coming from TIBCO this is reliable enterprise-class functionality.
The core technology is based on multi-core parallel processing that avoids the common pitfalls of bottlenecks and a failure to fully exploit parallel capability to provide the full potential of the technology. Currently they are looking at a capability in terms of a terabyte of memory across 32 cores, and all at affordable prices.
The technology is already proven to be more than capable of addressing such demanding applications as intra-day trading risk within the Financial Services market, and is now being applied to such applications as logistics, supply chain, and network management. I believe that rather than building bespoke solutions to the emerging needs of the Operational Intelligence market, Quartet FS offers an engine with all of the attributes required to offer reliable, and scalable performance, based on proven commercial of the shelf capability, making it a question of why would you not consider it as the most viable solution to your need?
We automatically stop accepting comments 180 days after a post is published. If you would like to know more about this subject, please contact us and we'll try to help.
Published by: electronicdawn Ltd.