Cloud

As more and more organisations are moving into petascale environments; driven by big data, unstructured explosions, day-to-day growth and generally poor data management; the ability to manage at scale is becoming increasing important.

Now from a vendor point of view, there has been a focus on getting to scale but that is less than half the story; if it is hard to implement and manage that scale, hard-pressed data-management teams are going to start looking elsewhere.  Managing at scale needs to be as easy and seamless as managing a single array or filer.

Implementation and expansion needs to be quick and painless; the ability to expand without little effort is a major show-stopper for many scalable implementations. I need to be able to add capacity to my systems to support I/O or data-growth but it needs to be transparent and non-disruptive; it needs to be automatic in its optimisation; quite frankly no-one has the time to re-layout a multi-petabyte environment manually with the almost inevitable disruption that brings.

Petascale computing almost always comes with a 24x7x365 availability requirement; Big Data analysis often involves long-running jobs.

But these huge environments bring other challenges as well; you will have large files, small files, tiny files, spread throughout your systems; access characteristics are different, some will be random, some will be sequential and in some cases, you might find both. Some files will have a single user and some will have hundreds of users. However, the data-management team will want to manage these all in a consistent and seamless manner; yet again, they will want to do this with a minimum amount of intervention.

Let’s think about the impact of a self-service environment where teams can throw up new environments; the data-management team will have little control of the files and type of files that these applications create. The provisioning tool may ask questions, will you produce large files or small files but in an agile environment, the answer given yesterday may not reflect the reality of the code written today.

This all leads us to a key requirement and feature for anyone who wants to sell Petascale data-management and storage tools; ‘Simple Scalability’. Yes, it is important that it is fast but it is equally important that it is simple to support and manage throughout its life-cycle.

Lets not kid ourselves; as we move to petascale and beyond; these environments are going to life-spans which far outstretch those of our current SAN environments because the practical realities of migrating petabytes of data stored in single system being accessed by many services is going drive this.

So the next time you are benchmarketing a system; ask yourself, is really practical or is it just a ‘My Dad is bigger than your Dad’ playground argument?

So in the latest little wing-ding between EMC and NetApp and who can do the fastest lap; I do wonder if they miss the point some-what. Benchmarks unfortunately generally focus on one thing, who can do ‘x’ faster than the competition; this is especially true of storage benchmarks which seem to throw up all kind of marketing monstrosities.

The problem with this is that life is not often that simple and performance is just one factor when purchasing storage.

I do wonder if the benchmarking industry could do with taking something out of Top Gear’s book and have ‘A Star Configuring An Extortionately Priced Array’; we could get a random star who has a book/film/album to promote and get them to configure an array to carry out a specific task.

The measure would then not only how well the array runs but also how long it takes them to get it to first I/O.

And I can see a whole series; ‘A Star Configuring An Extortionately Priced Private Cloud’ or perhaps ‘A Star Configuring An Reasonably Priced Public Cloud (just make sure that they’ve read the small print)’.

Whilst NetApp and EMC scrap over meaningless willy waving; the rest of the world try to get on with real work and real world problems. Yet again, sat in a meeting where the requirement is for a grow-forever and delete nothing archive but with the requirement for ‘instant retrieval’ and with a really simple user interface so that a non-technical user can cope. In fact they would really like the archive just to appear as a drive letter on their desktop and be able to just drag files back.

There is part of me who fancies doing a mad science experience using a bizarre mix of an Avere NAS accelerator and NFS-exporting an entire tape library using LTFS. I do wonder if it’d work…build a tape-based Dropbox like solution. Perhaps Avere fancy building a proof of concept system and taking that to NAB or something, I think it might get some attention.

Of course we’ll probably do something sensible and more conventional but a tape-cloud…that’d be kind of fun!!

There’s a lot of discussion about the resurgence of DAS and alternatives to SAN and NAS; whether these be virtual appliances, clustering storage, object or just plan old direct-attached-disk; all of these are seen as ways to replace the expensive network storage be it SAN or NAS attached.

But is this actually important or even especially new? It is certainly the case that the software vendors such as Microsoft and Oracle would like a piece of the action but we also have new players coming in via the virtualisation space.

Personally I see it just as another evolution in the realm of Networked Shared Storage; SAN, NAS and Clustered-Storage. The clustered storage will generally be built around commodity disk but it will not be exclusively so; it may be accessed in a variety of ways, clustered file-systems such as StorNext, GPFS and Gluster will all provide a block-level access but you can also throw in object technologies such as Caringo and you may still decide to access via the traditional NAS protocols.

There are certainly some interesting possibilities where block and file access could be provided to the same data; build yourself a storage-cluster and add in client nodes which see the storage as ‘local filesystems’ but also have remote access via NAS/CIFS or even object.

But is this really a resurgence of DAS? Not really, it’s still networked storage but just different. Existing SAN infrastructures can be leveraged to provide access to the physical bits (the rise of SSDs means no more storage is rust!). We simply have a new (actually old) tool in the box.

And it just reflects what we already know; Storage is Software…