Mike Shapiro is an expert in disruptive technology; he was working on Solaris in the early 2000s. He spoke to a number of us at Sun’s Guillemont Park Campus about Amber Road, Sun’s new disruptive file server technology. Sun and our customers have the opportunity to take advantage of the next big thing in network storage.

Mike explained that for a technology to be truly disruptive

  • it must be cheaper
  • it must be good enough
  • there must be a compelling reason for adoption

and Amber Road has two killer apps. Flash and Analytics. There is a small layer of functionality that Amber Road can’t provide but the bulk of the market doesn’t need it, and certainly doesn’t need it at the price charged. Since storage is a trust business, Sun’s storage sales teams and the customers need to understand very carefully the storage requirements. It is unlikely that any functionality not available is a universal requirement but in some cases, its not the right time for customers to move from their incumbent suppliers; they need some of the missing functionality. Talking to storage users about Sun’s new storage concentrates the minds of everyone involved.

Over the last 10 years, there have been only two ways that the laws of physics and economics permit to make disk arrays faster, either increase the cache size, or increase the disk speed. The cost of Flash has dropped over the last three years, thanks to those of us buying mp3 players and pdas. The Amber Road box’s software allow newly economic flash to do either or both. Sun is a leader in flash and certify enterprise flash for 3-5 years. and has additional advantages including the superior reliability of ZFS and the open source pricing of the Unified Storage arrays. We don’t licence a right to use. What we charge is based on what we ship, you don’t get charged more as you turn on functionality. (This has nearly always been true of Sun, I remember when buying SunOS systems that one of the advantages was that network functionality was bundled with UNIX where as I was asked to pay extra for networking and RAID functionality by my then incumbent supplier). Crucially Sun doesn’t seek to tax its customer’s innovation. The “no more to pay” approach also applies to the Analytics which come with the box and you can use them all.

Some of what is argued to be missing is FCAL support. Mike stated that the long-term winning strategy is to have only one cable going into the box. If there’s to be only one winner, it ain’t going to be FCAL; it needs to support Ethernet, and there’s a demand for infiniband. Our proposed iscsi functionality release plans means that the Unified Storage boxes can offer block devices over the network and support for most enterprise data centres will only get better.

The value proposition for Amber Road is that its cheaper, good enough and offers game changing superior management. This often gets lost in a feature benefit analysis, which often seek to disguise what the features cost. Sun knows storage and can meet the trust requirements that customer’s require, Amber Road shows that a trusted source can disrupt the economics, and its only the customers that win.

ooOOOoo

This was uploaded on 28th May 2009 and back dated to the date of occurrence, 11th March, on my sun/oracle blog, the was reposted on this blog in July 2016, comments about the simulator and try befoe buy were deleted due to irrelevance by 2016.

Squaring the circle, from disruption to trust
Tagged on:                             

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: