In the past I’ve attempted to write a single post for each vendor as a preview to Tech Field Day.  This time at Storage Field Day 11, for brevity, I’ve decided to divide the presenters into two groups.  Trying to find some connection between the vendors isn’t easy, but there is some bizarre logic here.  In this post I connect Avere Systems, Primary Data and Scality, all (start-up?) vendors with data management solutions.

Avere Systems

I’ve spoken about Avere Systems many times over recent years.  The company started out developing an edge filer to cache file content stored on data centre core NAS devices.  An edge device was/is effectively another appliance or virtual machine in a different data centre that locally manages NFS access and maintains a cache of frequently accessed content.  The solution is great for organisations that don’t want to replicate data between devices or deal with having to build out backup and data management capabilities in locations like branch offices.  However the solution was always planned to be more powerful than that and now allows data to be moved in and out of public clouds.  This means organisations can use public cloud compute to crunch their (file) data without having to go through wholesale migrations to achieve it.

Check out this recent post for a little background.  Avere recently announced their own storage appliance the C2N, which combines NAS and object storage with storage resources in the public cloud.  I expect the SFD presentation will focus on this new platform and what it has to offer enterprises looking to reduce costs and be more flexible with their data storage.  One other little thing; Avere managed to deliver very strong specSFS2008 benchmark results, despite a 150ms latency between one of their cache filers and backend S3 storage.  You can find out more details here (link).  This highlights how caching can work, if done right.

Primary Data

Expanding on the data management/abstraction theme, Primary Data is a start-up that has been working on a data virtualisation solution called DataSphere.  DataSphere works by abstracting and pooling the storage resources of an IT organisation.  These resources can then be mapped and accessed by hosts, with no involvement by DataSphere in the data path.  This is a different approach to previous monolithic systems that had trouble scaling because all I/O passed through a central set of hardware controllers.  DataSphere centralises metadata onto a redundant set of physical servers (which I believe can also be deployed in the public cloud).  Despite being out in beta for some time, DataSphere only went GA at VMworld this year, so I expect the presentation at SFD will be an update on the product and details of any changes made for the GA release.


Scality is a software-based object storage solution that also has support for file and block protocols.  The Scality platform is known as RING and this also describes the protocol used to manage communications between many nodes in a RING cluster.  Object storage has become both a category in its own right (with lots of initial uses around archive and backup) and a platform onto which other protocols can be built and deployed.  From a data storage perspective, object stores have always seemed to me to be the most “pure” way to store data on physical disk, although depending on how this is achieved, may not be the best from a performance perspective.  By this I mean that object storage platforms can be high performance, but serving (for example) block LUNs from an object store, may not be optimal.

Getting back to Scality for a moment, the company has expanded over time to add support for OpenStack, NFS/SMB as well as traditional object protocols.  Most recently Scality announced their free S3 Server, a node.js implementation of the Scality platform that runs in a Docker container.  This is a great way to test and validate the Scality platform and can be up and running within a few minutes.

The Architect’s View

This group of vendors are all focused on data management, in each case with the majority of their IP/intelligence in software.  Increasingly, managing data is becoming the important part of storage, not the physical hardware – although we’ll have more on that in the second group of presenters tomorrow.  In the meantime, you can find links to more content as a preview of the SFD11 presenters.

Further Reading

Comments are always welcome; please read our Comments Policy first.  If you have any related links of interest, please feel free to add them as a comment for consideration.  

Disclaimer:  I was personally invited to attend Storage Field Day 11, with the event teams covering some of my travel and accommodation costs.  However I was not compensated for my time.  I am not required to blog on any content; blog posts are not edited or reviewed by the presenters or the respective companies prior to publication.  

Copyright (c) 2009-2016 – Chris M Evans, first published on, do not reproduce without permission.

Written by Chris Evans

With 30+ years in IT, Chris has worked on everything from mainframe to open platforms, Windows and more. During that time, he has focused on storage, developed software and even co-founded a music company in the late 1990s. These days it's all about analysis, advice and consultancy.