Big Data upstart Iguaz unveils furiously complex exercise to remove complexity

Startup Iguaz.io has unveiled a virtual (big) data services architecture which it says can deliver data in the needed firms to apps in the Big Data infrastructure without multiple and repeated data copying and conversion.

The company is a 2014-founded, $15m A-round-funded startup. Its pitch is that the entirety of the data supply workflows in the Big Data sphere are over-complicated and represent band-aid on band-aid.

What it says we need is something [cue marketing-speak] that consolidates data into a high-volume, high-velocity, real-time data repository design that virtualises and transforms data on the fly, exposes it as streams, messages, files, objects or data records consistently, and stores it on different memory or storage tiers.

Here is a diagram it has supplied, mentioning Hadoop, HBase, Spark, Splunk, Kafka and Redis, and showing the complexity which it says is endemic:

This is its high-level architectural design to fix the mess:

The idea is roughly similar to that of Actifio and Primary Data in that there is a single copy of data, spread across tiers, which is used to provide data in the right form for applications that need it. This is, by the way, a fantastically complex thing to do.

Iguaz claims this kind of scheme can offer enterprises a 10x-to-100x improvement in time-to-insights at lower costs, which is easy to say as there is no product and no standard Big Data pipeline flow time.

There is no product yet but Iguaz has 40-plus employees. What we have here, we believe, is that Iguaz has come up with a high-level software product design, which could be helpful in getting funding for subsequent development. It has to develop v1.0 technology and demonstrate that. If it follows the traditional VC funding route this will then be used to unlock further funding and produce a product that can be tested by potential customers.

Watch this space. Iguaz is setting out on a furiously complex exercise to remove complexity from the sprawling Big Data store and analytics data workflow mess. The next step is a technology demo and we look forward to that. ®