Storagebod Rotating Header Image

Storage

Clue is in the Name

Dec 11th, 2011
by Martin Glassborow.
No comments yet

Information Technology; that’s what we do and of those two words surely the most important is Information.

I like Gregory Bateson‘s definition of Information as ‘Difference which makes a difference’.

As we start to move into the realms of ‘Big Data’; we should remember what we are really trying to do, we are not analysing ‘Big Data’ to boast about the size of our data-sets, how much data we are analysing; we are looking for information, the ‘differences which make a difference’ to our companies, our lives and the general world.

We should also be concentrating on trying to do this as efficiently as possible; what technology is enabling us to do is to analyse ever bigger samples in ever shortening time-scales but it does not always correlate that the bigger samples will enable us to spot ‘Differences which make a difference’ any better.

Make sure the driver for better technology is Better Information and not simply better technology. The siren song of bigger is some which could smash us onto the rocks; if someone can make the same decisions with less and do it even quicker, the competitive advantage can be lost.

I personally would have been much happier if the ‘Big Data’ meme was actually ‘Better Information through Technology’, not as catchy maybe but more relevant and real in my opinion.

But don’t get caught up in the Cult of Big! Remember it’s not all about the Data, it’s about the Information.

Solutions and Specialists…

Nov 21st, 2011
by Martin Glassborow.
No comments yet

Solutions are great, a vendor turns up and sells a turn-key solution, it’s a marvellous world and everything just works; it’s all certified and lovely. There is a single supplier to kick and trouble-shoot the problem. Or at least that’s the theory….

But what happens when the supplier can’t fix the problem? Who do you turn to then? Funnily enough, that’s the situation I’m in today. A turn-key media solution which we’ve been kept at arms length from for years has developed issues and now who do the customer turn to when not getting good answers from the solutions vendor? That’s right, our little team of specialists..

A couple of hours of investigation and a meeting with the vendor where we expose what we see as issues by treating the solution as just another piece of infrastructure has been enlightening to both our internal customer and the solutions vendor. There will be no arms-length going forward; solutions still need specialists.

You need specialists to ensure that the wool is not being pulled over your eyes; you need specialists to ask the right questions and know when the answers are not good enough. Too often you will find that the solutions vendor themselves have little clue about the underlying infrastructure; focusing on the application whilst using the hardware as a nice little revenue lift. This is fine until you hit problems.

If you are buying integrated solutions and stacks; make sure that they are integrated and that the solutions provider can actually support the stack. Don’t be afraid to dig into what is being provided as part of the stack/solution and keep some specialists around.

 

Simple Scalability

Nov 17th, 2011
by Martin Glassborow.
1 comment

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?

 

A Star Configuring…..

Nov 14th, 2011
by Martin Glassborow.
3 comments

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)’.

 

Always Questions

Nov 8th, 2011
by Martin Glassborow.
1 comment

Matt Davis touches on a number of my favourite storage subjects in his post here, so I thought I’d take it and run with it a little bit with some statements and some questions.

Firstly, the storage array is becoming software layered on top of a commodity platform.

Secondly and related, hardware is becoming less of differentiator between the array types from all vendors.

Thirdly, if the only differentiator is software; why really do EMC have VNX and VMAX? Because they do I suspect is the real answer and glomming those two lines together is the job of the next guy.

Fourthly, why do I really pay so much extra for VMAX? What are the relative sizes of the two development teams and is VMAX software so much better? Or is it that to make it really Enterprise, you have to charge more?

Fifthly, could EMC build a multi-head VNX to support block? Could they build a multi-head VNX to support file?

Sixthly, would not the ideal EMC platform be VMAX and Isilon running on the same hardware to provide scale-out block and file? If you feel brave, perhaps run Atmos to support Object as well?

Seventhly, do EMC *really* need Cisco to produce the next generations of vBlock?

A few statements and more questions. I find EMC a fascinating company; there is the appearance of clarity of vision but at times, you have to wonder if there is true clarity of purpose.

Thanks Matt for the inspiration..

Features /= Product

Nov 6th, 2011
by Martin Glassborow.
No comments yet

Hey I’ve got a great idea for a new storage product often equates to ‘I’ve got a great idea for a new feature’; companies large and small are equally guilty of this but a product is more than just a feature. Wandering round SNW and IP-Expo, this really struck me; lots of people are basing their whole product around a single feature.

Unless you have managed to invent a completely new product category; for example NetApp with their NAS appliance was really a new category (ignoring Auspex), you need more than a great feature and you need to do a whole lot more. 3Par for example may have been known as the thin-provisioning company but their product was more than just thin-provisioning; they lead the way on ease of use, wide-striping (okay, the enabling technology for thin-provisioning), micro-RAID and customer engagement.

If you just have a feature and it is something which could be easily be integrated into an existing competing product, you are going to have to make the most of your first mover advantage and then possibly get out. This may seem obvious but I am seeing a number of companies at the moment who are doing a feature and do not currently seem to be progressing whilst there are oncoming juggernauts who could possibly crush them like a bug.

And there are very few new ideas out there which are not being worked on by at least three different companies.

Mad Science Experiments

Nov 4th, 2011
by Martin Glassborow.
5 comments

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!!

 

SNW Europe – Some Thoughts

Nov 4th, 2011
by Martin Glassborow.
5 comments

I had an interesting chat with Bob Plumridge of SNIA whilst at SNW-Europe where we discussed the differences between the European organisation and the US organisation; possible directions for the Academy events and SNW-Europe and what we want to see from the industry organisation.

There are significant differences between the US and the European SNIA organisations; SNIA in the US is a fairly inward looking organisation, focusing very much on the development of standards and discussion between vendors, whereas the European organisation is generally more outward looking and focused on the education aspects. Educating end-users about technology concepts and trends but in a way which does not focus on one vendor or another.

I’ve only ever attended SNW Europe and not its sister conference in the US but the comments I have heard from people are that the European event gets a lot more end-users and is less of a industry networking event. This should mean that it has more value and attendance is certainly up.

But it may be time for it to change again or at least evolve so that it maintains value. Here are some of my thoughts which I shared.

1) Education sessions seem to be aimed at a relatively low-level of assumed knowledge. There might be scope for a more advanced track.

2) Perhaps cover areas which although not traditionally storage but do have applicability to us; for example the legal obligations with regards to data security and especially with regards to locality of data when storing in a cloud infrastructure.

3) Birds of the Feather sessions; where like minded storage folks can talk about the things which are impacting them and their businesses.

4) Ensuring that we have good attendance from some of the more interesting start-ups.

The vendor neutral presentations are good but they only go so far; I do wonder if it might be allowable to let the vendor who has just struggled through a presentation to either take questions and not having to be completely neutral in their answers or allowing them a five minute lightning pitch at the end of a presentation.

Or perhaps have a session which is just lightning pitches? Or a vendor balloon-debate?

I think as an event, it’s good but at times it takes itself too seriously. It could perhaps do with a fun side as well…what would you like to see?

 

Fabric Causes Over Excitement.

Nov 2nd, 2011
by Martin Glassborow.
1 comment

There is currently a fair bit of noise about FCoE, whether it is ready for the prime time, whether it’ll ever be ready and does it matter anyway. I thought I’d add my tuppence worth of thought.

When FCoE was first announced and it looked like that we might get some fabric convergence, I was pretty positive about it all. I really thought that this was a good thing, convergence is good. Now there have been various ways of getting storage traffic to flow over traditional IP networks but storage guys don’t trust them and quite frankly most network guys don’t trust the storage guys not to destroy their networks anyway.

FCoE and the accompanying standards seem to be a real chance to do something about and as we continue down the path of 10GbE in the data-centre; we could work together to bring a new world and operating model. (exaggeration for effect BTW)

However, we are a long way away from this at present and 10 GbE networks are going without consideration of FCoE and traditional SAN environments keep growing. The vendors cannot agree on how to ensure interoperability and standardisation. So FCoE is turning into the complete ‘head-f**k’ that FC is. You will not be able to easily build a heterogenous FCoE fabric, mixing and matching switches from different vendors; you will have a ‘vendor x’  fabric, you will still be in the world of complex compatibility matrices, you will have political trouble as you try to get teams to co-operate and work together, you probably will not see any real performance gains at present from going down the FCoE route and change/problem/incident will be a complete mess.

So why bother with FCoE at all? You could just run iSCSI/NAS over 10GbE or just stick with FC as you know and love it today. The roadmap for FC is still healthy and there is deployed product. You probably have enough going on in your data centre already and if IT is going through a transformational change…you really do have enough on your plate.

Now, the future is not completely bleak for FCoE but the adoption of FCoE is going to take longer and more painful than I thought. But the vendors do need to get together on work out how best to adhere and stick to the standards and not just simply ‘Embrace and Extend’ causing a reply of FC!

Brief Bod

Oct 21st, 2011
by Martin Glassborow.
No comments yet

So I’m going to be at SNW Europe (all funded by the nice people who run it) and part of the deal is to spend time with vendors and let them brief me about their products; it’s almost like being a real journalist or analyst but of course I’m not!

And this seems to make it harder to know how to engage with me and with my fellow bloggers. Vendors try to work out whether to sell to us as many of us are end-users in our day jobs or do they treat us like they would a more traditional journalist.

So here’s some thoughts…

1) Read the blog; you don’t have to read the every entry but you should have an idea of what interests me and my style. We all have different styles; I don’t blog that much about product at the moment, so spamming me with speeds and feeds does not really work. I like to think that I blog more about themes and memes; thinking about why we do something as opposed to the detailed how.

2) Don’t talk at me; be prepared for a conversation. I’m not simply going to take your word for anything. As a end-user, I will use my experience and experiences as reference points but this is not an excuse for you to try to sell.

3) Don’t expect me to write a post about your product but you may find that it gets a mention in a broader piece about a theme or idea. I blog for ‘fun’ and not for profit…

4) Relax and enjoy it….us bloggers are generally a friendly bunch and the fact that we do this for fun means that it should be fun for you too. I’m not looking for a scoop or an exclusive; I’m just looking to find out cool stuff!

So tell me something cool….

 

← Earlier Posts Newer Posts →