Storagebod Rotating Header Image

October, 2011:

Innovate At Your Peril

Oct 22nd, 2011
by Martin Glassborow.
5 comments

This article was linked to by some of the usual EMC suspects; a fluff and puff piece about Private Cloud with the normal warnings about security in the Public Cloud. It is this section of the article which I find especially disturbing both in tone and message…

I’ll leave you with what has become my favorite story and it was told at CIO 100: Apparently, two engineers at a pharmaceutical company had to complete a critical project quickly and bid it out to IT. IT came back with a massive cost and a timeline in months. The engineers instead used their credit cards to use cloud services and completed the project in a few weeks and won an award for cost savings. The day after winning the award, both were terminated for violating the firm’s security policy as the project, which was ultra-secret, hadn’t been adequately secured.

I can almost imagine the teller of the tale’s gleeful smile as he recounted that story, perhaps the CIO involved. Now I think there should have been several different actions, none of which lead to the dismissal of two obviously talented and thoughtful engineers.

1) The CIO should have been hauled up and made to explain why his team could not provide the services that the engineers needed in a cost effective and timely manner. He put them in the position that do their job properly, they had to bend the rules. In fact he should be the person loosing his job and as a result of his inability to provide service; the company had had to terminate two valuable employees.

2) The team which looks after security should have been asked to look at the project and what the engineers had done; make a proper security assessment and work with them to ensure that such projects could be delivered in the Public Cloud in a secure manner. Proper procedures and guidelines should be put in place to support innovation.

But instead, a vengeful IT department decided that best thing to do is to shut down anyone innovating in their space.

And if anyone thinks that the large pharmaceuticals are not using public cloud; you should probably think again. They are regularly and I suspect securely; or perhaps, its not 100% secure but the opportunity for quicker delivery is worth risk.

Security is an issue but don’t let vendors and IT departments use it to block innovation and keep their castle intact. Security needs to move on from ‘No!’ to ‘How can we help you achieve your goals!?’; a bit like IT departments in general.

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….

 

Storagebod’s Fashion Tips for Storage Engineers

Oct 17th, 2011
by Martin Glassborow.
1 comment

Greg recently wrote his Fashion Tips for Network Engineering Men and in doing so, I think he has highlighted some of the differences between Network Engineers and Storage Engineers. So I thought I would take the challenge of giving the sartorially challenged Storage Engineer some help.

Fortunately, I have some fashion pedigree being related to both the Duchess of Cambridge and the Head Buyer for Harvey Nichols; in fact of the three of us, I suspect I am the true fashion leader in the family.

Ties

Every Storage Engineer needs a good collection of ties; these should rarely be worn in the conventional manner but at times, they come in handy for keeping sweat out of your eyes whilst trying to explain why Fibre Channel is a far superior protocol to anything that a Network Engineer might use. Indeed, a number of them can be deployed to prevent said Engineer escaping.

The exception to this is the Black Tie; you should have at least one which can be worn to the many storage award ceremonies which you might get invited to. These are deeply meaningful events where awards are given on a strictly meritorious basis. And they are not an excuse to simply drink large quantities of cheap champagne and wine.

Shirts

Vendor polo shirts should be worn whenever possible; these allow you to demonstrate what high esteem your vendor holds you in. However never let said vendor feel too comfortable, always wear a different vendor polo shirt to the one you are having a meeting with.

For days when you want to dress to impress, nothing says engineer like a dress shirt with the sleeves completely removed.

And for those dress down days, can I suggest a t-shirt with the sleeves removed? Never a vest, that simply looks tacky!

Hats

Face it, no-one looks good in a hat; if you need to keep the sun off your head (and that is unlikely, you are shut away from the sun most of the time), a knotted hankie should suffice.

Portable Computing

Portable computing is not for the Storage Engineer! If you can carry it, it’s non-enterprise kit and we don’t do non-enterprise. Wireless? Pah, if it’s not a fibre cable; it’s not connectivity!

Belt

Belts should be elasticated and not leather. They should be capable of being used as a catapult or if there are a number of you; a ballista.

For the particular daring of you; they can be used for bungee jumping after an awards ceremony.

A diving belt can be used as an albeit for carrying a diving knife but a diving knife is normally overkill for most occasions.

Suits

A dinner suit is the only suit you will likely require, to be worn to important awards ceremonies.

The old favourite boiler suit needs to be worn with care unless you end up making a political statement that you do not wish.

If you end up in court after an awards ceremony, may I suggest your birthday suit; it always impresses the court that you really have nothing to hide.

Shoes

A real engineer would never wear shoes; boots are the preferred wear. Site boots are especially effective wear when bringing complex vendor negotiations to a close.

Converse can be worn but never with a suit; you’ll just end up looking like to you want to be David Tennant in Doctor Who!

Socks

Preferably clean or at least the ones from the bottom of the laundry basket. It is one the place that I really agree with Greg though; simply buy a dozen or so pairs of black socks. That’s unless you are ignoring my footwear advice and are wearing shoes, then white socks are preferred. So much more stylish!

Trousers

Yes! In general trousers are a good idea! Although, a kilt can be an acceptable alternative.

Shorts, especially lycra cycling shorts just make you look like a cycle courier and you are better than that.

Be Amazing

If you take my advice, everyone will know you are a Storage Engineer and you will get the respect due!

#storagebeers – 20th October

Oct 17th, 2011
by Martin Glassborow.
No comments yet

Just a quick reminder, there will #storagebeers after IP-Expo on Thursday October 20th.

There are also #vbeers after IP-Expo on Wednesday.

Both events are happening at the same pub – The Atlas which is a stone’s throw from Earls Court.

The Atlas

16 Seagrave Road

London

SW6 1RX

I also know that there are various events happening in Copenhagen for VMworld.

So pop along to one of the events if you can.

Also remember, there will #storagebeers at SNW Europe in Frankfurt.

I look forward to meeting some of you in coming weeks.

500 Not Out! How Did This Happen?

Oct 13th, 2011
by Martin Glassborow.
2 comments

Another milestone for the Storagebod blog; this is my five hundredth post!  I should write something really meaningful but I’m not sure I can think of anything, so I thought I’d just put down a number of short ideas which might get developed more into full blogs.

Enterprise IT – Enterprise IT is a meaningless term; it is insulting and disparaging of anyone else who is not an Enterprise. If you run a business and your IT infrastructure is core to the continuation of your business; you need IT which is reliable, scalable and all those other good things. You can of course leave off paying the premium for what people call ‘Enterprise’!

RFPs – Request For Pain. RFPs generally exist for one reason in IT; that is to give a bunch of vendors a kicking. In the storage world, you probably have little reason to change vendor but you might as well kick the crap out of your incumbent for laughs. The result of nearly all RFPs is driven by politics and not technical reasons; it is probably better for your sanity if you acknowledge that up front. If a customer really wants to change, they will.

It’s a PC Plus World; the reality is that if you’ve got a desktop, you will probably keep a desktop. Don’t expect this to change any time soon. Yes, you will probably be able to get access to some services via an alternative device but I suspect that most desktop users will stay just that.  You will see more mobile devices about but we all know that it’s a pose and it gives us something to do in tedious meetings.

Big Data; use Big Data to make better decisions, don’t use it as an excuse to dive into analysis paralysis. If it has all the characteristics of a duck, it probably is a duck…you don’t need to decode it’s genome before you serve it with Orange sauce.

Cloud; it’s a way of delivering service but it’s not the only way of delivering service. If you find yourself getting religious about how you deliver a service as opposed to delivering the service…take a holiday and get some perspective.

Internal Service Providers; you only have one customer to focus on. This is your biggest strength and weakness.

IT Management; take the chance if possible to manage a team outside of your technical experience. You learn to manage, delegate and trust your team; you focus on managing and not trying to do two jobs. You can always go back to managing in your technical discipline but you will bring new insight and ideas.

Work/Life Balance; you will die, this is inevitable. Make sure that the people you love remember you for the right reasons and not for times you weren’t there.

So that’s post 500 done…here’s to the next 500!!

 

Trading Commodities

Oct 12th, 2011
by Martin Glassborow.
3 comments

‘Would you trust your business on a storage array built from commodity hardware?’ to paraphrase a remark which came up in a meeting today? This comment took me aback as we were discussing another array which is also built from commodity hardware, although the questioner seemed blissfully unaware of that. I left meeting feeling a little perturbed and put out with something nagging going round my head.

The comment is not an uncommon one to be honest but does it really mean anything at all? And then it hit me,

‘We risk all of our business on commodity hardware all the time; what the hell do you think those servers are?’

Most of the time, they will be clustered to fail over in a very similar manner to a commodity dual-head storage device. And as we put virtualise more and more services; the impact of a failed server or server-chassis is possibly very similar to the impact of a the failure of a head in a dual-head storage array.

So would I trust my business to commodity hardware? Well, I don’t think we’ve got that much choice these days, do you? Be it storage or servers; its getting to be pretty much the same thing!

More Data

Oct 10th, 2011
by Martin Glassborow.
2 comments

One of the problems with Big Data is that it is Big; this may seem obvious but actually to many people it’s not. When they hear Big Data what they end up thinking is that is simply More Data; so they engineer a solution based upon that premise.

What they don’t realise is that they are not simply dealing with More Data; they are dealing with Big Data! So for example, I know of one initiative which currently captures, lets say 50,000 data points today and someone has decided that it might be better if it captured 5,000,000 data points. Now, the solution to this is simply to throw bigger hardware at it and  not to re-engineer the underlying database.

Yes, there’s going to be some tweaking and work done on the indices but ultimately, it will be the same database. Now, this is not to say that this will not work but will it work when the 5,000,000 inevitably becomes 50,000,000 data points? It is simply not enough to extrapolate performance from your existing solutions but how much of capacity planning is simply that?

If you are already in the position that you are looking at More Data; it will probably become Big Data before you know it and if you haven’t already engineered for it, you are going to have a legacy, dare I say it ‘Toxic’ situation in short order.

Everything is changing; don’t get left behind.

Big /=More; Big == Different.

Think Big, Think Different.

 

Toxic IT?

Oct 8th, 2011
by Martin Glassborow.
1 comment

Often we find ourselves talking about Legacy IT; especially when we are discussing the move to ‘Cloud’. What do we do with the legacy? At CloudCamp London, Simon Wardley has suggested that we need to start calling Legacy IT, Toxic IT. And the more I think about it, the more I agree but not just when are talking about the move to Cloud.

Many organisations have Legacy IT; oh, the accounts system; that’s legacy; the HR systems, they are legacy; that really important system which we rely on and is foundational, that’s legacy. What? These are all key systems, how have they become legacy? And the longer we leave them, the more toxic they become.

Businesses and especially Enterprises run on legacy systems; whilst we rush to the new and roll-out exciting services built on the new Cloud, we leave the legacy behind. And so they moulder and rot eventually become toxic.

All of us working in Enterprise IT can probably point to at least one key service which is running on infrastructure (hardware and software) that is long out of support. Services which may well be responsible to a large proportion of the company’s revenue or the company’s survival. These services have been around since the company was founded; that’s why I talk about them being foundational.

But why are they left behind, maybe it’s because it is easier to ask for budget for new stuff that brings new value and markets to a company? What is the return on investment on maintaining your account systems? It’s not going to add to your bottom line is it? Still, if your accounts systems collapses; you won’t even know what you your bottom line is.

So, that Legacy IT; it’s rapidly becoming toxic and it is going to cost you more and more to clean up that toxic pile the longer you leave it.

I think it’s time for many Enterprises to run an Honesty Commission where they ask their IT teams to identify all systems which are becoming toxic and commit to cleaning it up. Just because it hasn’t broke yet, it does not mean that it is not broken! Just because your services are not showing symptoms of toxicity, it does not mean that they are not slowly breaking down. Many poisons work rather slowly.

Yes, you might decide that you are going to move it to the Cloud but you might just commit to maintaining it properly.

The DAS Alternative?

Oct 6th, 2011
by Martin Glassborow.
1 comment

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…

Tiers are back

Oct 4th, 2011
by Martin Glassborow.
No comments yet

No sooner than we start get to technology which allows us to practically get rid of tiers or at least the management of tiers by the introduction of automation to move data around arrays with minimal admin intervention; we start to bring a whole new bunch of storage silos.

There seems to be discussion about tiers again and how we categorise data; so we have SSD only arrays which are our performance tiers, we have SATA/NL-SAS which are capacity tiers and then we have tape/MAID etc which are our archive tiers. So here we go again, fragmentation of data across multiple tiers could well be with us again with-in the next half-dozen or so years.

And there will be hand-wringing again; ILM and HSM is too hard say the vendors, the returns are minimal. Once we see the mass introduction of SSD-only arrays or perhaps that should be an if; I think as storage consumers that we need to be putting pressure on the vendors to provide seamless ways to move data around between the arrays.

I wonder if almost inevitably, the standard operating model will be external virtualisation in front of silos? Or is it application level intelligence which does the movement? Or the file-system?

I am not against the introduction of SSD only arrays; actually I think starting from a different point as opposed to just putting SSDs into an existing array is not such a bad idea. Designing from the metal up may mean we see less compromises and more effective use of SSDs, yet it does bring challenges into the data centre.

At times, this all feels very circular.

← Earlier Posts