Storagebod's Blog

Interestingly Un-Interesting

3Par finally get acquired and by Dell; mission accomplished by their bridgehead-man, Pathfinder Farley. Like love and marriage; Farley and Dell go together like a horse an carriage. 

But all joking aside, what does this really mean.

It does not mean that Dell have an enterprise storage strategy; just having a bunch of SKUs does not make a strategy. Over the past twelve months or so tho', Dell have acquired some interesting technologies.

1) Exanet - a scalable NAS product.

2) Ocarina - content aware DeDupe technology.

3) 3Par - an Enterprise Storage company.

Some people may raise eye-brows at the last statement but I believe that 3Par were really an Enterprise Storage company; or at least they had aspirations to be so. From pretty much day one, they set their sights on HDS/EMC/IBM/HP's Enterprise market and considering how hard it is to compete in that market, they did remarkably well. To gain any kind of footprint at all was an achievement and from those customers who bought them, I heard only good things. People liked 3Par, both the technology and the company. They took a difficult road and who is to say that they were wrong? The investors have got a pretty exit out of this. 

But what have Dell got out of this? They've got some good technology and some great goodwill. Can they make anything out of this?

1) Exanet - forget the NAS, it's all about ExaFS. Massively scalable, clustered file-system. 

2) Ocarina - Intelligent DeDupe which can deal with multimedia files.

3) 3Par - Enterprise Array with pretensions to federation, predictable performance and management simplicity.

Throw in Equallogic and you've got a storage cloud; it's not a especially interesting storage cloud in that it doesn't really do anything that new but if Dell can keep it very simple, they may have product as opposed to a bunch of technologies. Does it push the boundaries of what Enterprise Storage is about? No, not at all but this is Dell we are talking about and Dell are not about interesting but commoditising. 

All quite boring but maybe in a good way.

The Late Storage Industry!

The Storage Industry is Late; no not dead but late. So instead of picking on just a single person, I'm going to call-out a whole load of vendors.

1) FAST2 on VMAX/DMX is late; the announcement was over a year ago and still there is no public GA announcement. No wonder Gelsinger is upping the amount of spending on R&D but I'd prefer him to spend money on Research, Development and Delivery.

2) OnTap 8.0 was late and despite all the current mutterings from the NetApp cabal; do we really expect the subsequent releases to be delivered in any more of a timely manner? OnTap 8.0 was feature light; a holding release and needs to get to 8.0.1 quickly and 8.1 soon after. I don't believe that NetApp have the resources to maintain multiple code trees, they need to get as much of their install base to 8.x as quickly as possible if they are to concentrate on innovating. 

3) USP-NG or whatever it is to be called is late! I think a lot of us were expecting the replacement to be announced pretty much at the time of VMAX, yet we heard but the gentle chirping of crickets. 

4) XIV enhancements are late; much of the NDA stuff I was told about two years ago has yet to be delivered. This was stuff which was supposedly easy to deliver and now it's just plain late. Instead we hear of a whole new bunch of features which will probably be late as well!

5) Soracle's delivery of a storage strategy is so late that it's looking like a Norwegian Blue.

6) EVA replacement? Is it late or just dead? Who knows?  

I'm sure there are many other examples of tardiness in the industry! 

If you feel left out and want to tell me which of your products that you are late with, I'd be happy to listen! 

Managing Migration Makes Martin Mad!

So we can thin-provision, de-dupe and compress storage; we can automate the movement of the data between tiers; now one single array may not have all these features today but pretty much every vendor has them road-mapped in some form or another. Storage Efficiency has been the watch-word and long may it continue to be so. 

All of these features reduce the amount of money that we have to pay for our spinning rust; this is mostly a capital saving with a limited impact on operational expenditure. But there is more to life than money and capital expenditure; storage needs to become truly efficient through-out its life cycle; from acquisition to operation to disposal. And although some operational efficiencies have been realised, we are still some distance from a storage infrastructure that is efficient and effective throughout its life-cycle.

Storage Management software still arguably is in its infancy (although some may claim some vendor's tools are in their dotage); the tools are very much focused at the provisioning task. Improving initial provisioning has been the focus of many of the tools and it has got much better; certainly most provision tasks are point and click operations from the GUI and with thin and wide provisioning, much of the complexity has gone away. 

But provisioning is not the be all and end all of Storage Administration and Management and it is certainly only one part of the life-cycle of a storage volume. 

Once a volume has been provisioned, many things can happen to it; 

    i) it could stay the same

    ii) it could grow 

    iii) it could shrink

    iv) it could move within the array

    v)  it could change protection levels

    vi) it could be decommissioned

    vii) it could be replicated

    viii) it could be snapped

    ix) it could be cloned

    x) it could be deleted

    xi) it could be migrated

And it is that last one which is particularly time-consuming and generally painful; as has been pointed out a few times recently, there is no easy way to migrate a NetApp 32-bit aggregate to a 64-bit aggregate; there is currently no easy way to move from a traditional EMC LUN to a Virtual Provisioned one; and these are just examples within an array. 

Seamlessly moving data between arrays with no outage to the service is currently time-consuming and hard; yes, you can do it, I've migrated terabytes of data between EMC and IBM arrays with no outage using volume management tools but this was when large arrays were less than 50 Tb. 

We also have to consider things like moving replication configuration, snapped data, cloned data, de-duped data, compressed data; will the data rehydrate in the process of moving? Even within array families and even between code levels, I have to consider whether all the features at level X of the code are available at level Y of the code. 

As arrays get bigger, I could easily find myself in a constant state of migration; we turn our noses up at arrays which are less than 100 Tb which when we are talking in estates which are several petabytes is understandable but moving 100s of Tb around to ensure that we can refresh an array is no mean feat and will be a continuous process. Pretty much once I've migrated the data, it's going to be time to consider moving it again. 

There are things which vendors could consider; architectural changes which might make the process easier. Designing arrays with migration and movement in mind; ensure that I don't have to move data to upgrade code levels; perhaps consider modularising the array, so that I can upgrade the controllers without changing the disk. Data-in-place upgrades have been available even for hardware upgrades; this needs to become standard. 

Ways to export the existing configuration of an array; import it onto new array, perhaps even using performance data collected from existing array to optimise layout and then replicate the existing array's data to enable a less cumbersome migration approach. These are things which will make the job of migration more simple.

Of course, the big problem is.....these features are not really sexy and don't sell arrays. Headline features like De-Dupe, Compression, Automated Tiering, Expensive Fast Disks; they sell arrays. But perhaps once all arrays have them, perhaps then we'll see the tools that will really drive operational efficiencies appear.  

p.s I know, very poor attempt at a Tabloid Alliterative Headline

And Bod Makes Three

It is funny that whenever you write negative about NetApp, it is often that accusations of bias fly. I will admit that this rarely comes from the NetApp bloggers but there does appear to be vociferous chorus of what could be called 'Fanbois'. I can pretty much write whatever I want about EMC, HDS, HP or IBM (with the exception of XIV) el al and expect the criticism to be taken on the chin or at least rebuffed in reasoned and reasonable terms.

NetApp have a great product; they changed the NAS storage market, indeed arguably they created the market. Yes, I know I am ignoring Auspex and unlike many, I actually used an Auspex but it was NetApp which took the market by storm. NetApp are the currently deservedly the market leader in this space but I would suggest that currently that their technical lead is slowly been eaten away. The leader is there to be shot at and the bullets are flying. 

My good friends Chris and Ian have written a couple of blog entries; a bit critical but more importantly they ask some questions of NetApp. I think there are more questions to be asked. 

I cannot be the only one who is wondering why we have not seen a major hardware refresh in the Filer range; OnTap 8 probably drives the hardware harder and I do wonder if some of the restrictions around ASIS volume sizes are simply down to lack of sheer grunt in the existing Filers?  

Although we often say that the hardware is no longer important and it's all down to the software; it is important that the software runs on the best possible and most cost effective platform. The more grunt you have, the more that your software can do. And it is surprising that we've not seen a hardware refresh as it should be relatively easy to do but I suppose keeping the Filers on 'old' technology improves margins for NetApp. 

And as a unified storage platform; the vSeries especially could do with being able to support more ports/interfaces. 

As well as the hardware, there is also the Bycast acquisition; this area has gone particularly quiet and pretty much into stealth mode. In fact, I challenge you to find a link to  on the NetApp front page? It's actually hidden under Products >> Management Software; the same category as Sanscreen. In fact, Sanscreen is not something you hear a great deal about at the moment either.

EMC got complacent and let HDS and then NetApp make hay in their core markets. VMware gave NetApp a huge-in by advocating NFS when EMC's products were not strong in that place. It's very easy to get complacent when you are making money hand over fist and when at times it feels like your competitors are doing their level best to trip up.

Storagebod likes this!

I really like this blog entry by Vince Westin; it absolutely confirms a belief that I have held for sometime! Application vendors have shares in storage companies! 

Poorly written SQL statements have probably driven more high-end array purchases than any salesman working for a vendor. Actually I suspect that poorly written SQL statements have driven more Enterprise IT purchases than anything. 

It is actually nice to see a vendor rather than recommending an expensive upgrade actually doing the sensible thing and looking for the underlying problems. 

Still perhaps Oracle purchased Sun because they absolutely realise that this is the case; they can now sell you both a nut but also a very expensive sledge-hammer to crack the nut. 

I suppose if I have one big concern about Cloud is that if we allow people to treat resource as abundant and cheap enough to waste; will code-bloat and poorly written code proliferate?

 I am convinced one of the major drivers of storage growth is that desktop storage is now so cheap; it can be wasted but the same is not yet really true of Enterprise Storage; a terabyte of 'Enterprise SATA' is very much more expensive than a terabyte of 'Desktop Storage' but try telling the average developer or user that? 

With Abundant Good Cheer

As we move from a time of IT as a scarce and controlled resource to a time where IT is seen as an abundant and easily available resource; we need to consider what this means to us in Enterprise IT. Is abundance a good thing and a power for good or does it bring with it issues? 

Lets look at a couple of quotes from Clay Shirky's book 'Cognitive Surplus'

'Abundance breaks many more things than scarcity does. Society knows how to react to scarcity' 

'Abundance is different: its advent means we can start treating previously valuable things as if they were cheap enough to waste, which is to say cheap enough to experiment with.' 

Now Shirky is talking about publishing, information and how we consume/create media but some of his thinking strikes a surprising amount of resonance as to what is going on in Enterprise IT today and how it is beginning to change.

In Enterprise IT, we know how to control access to resource, how to ration resource in such a way to ensure that we always have some resource saved and preserved for a rainy-day. But as we move to a world where resource is abundant; our models have to change. 

Perhaps we need to move from beyond the 'gate-keeping' mode of operation that many of us are used to where our job was only to let people into our environment if they met our very stringent criteria to a mode of operation more akin to one of the accommodating host where everyone is welcome as long as they do not impact any of the other guests.  

Enterprise IT is probably not yet at the stage where it can be considered to be cheap enough to waste but perhaps we are on the cusp of this? Or perhaps we need to be considering what needs to be done to get to that position. The Amazon EC2s of this world certainly allow experimentation at a very low cost financially. And arguably more important, the time to implement is much quicker. 

This may drive a culture of 'try often, fail a lot, succeed rarely'; as in the world of publishing, the average level of application quality may well be driven down and it will be important that we build infrastructures which can cope with failures. Badly behaved applications will need to be managed, we will need ways of spotting a badly behaved application quickly and like the good host, gently encourage it to behave better with the eventual sanction of kicking it out. 

Virtualisation can help to ensure that applications behave within limits but applications can fail in many ways and we will need to ensure that we have the tools to tidy up afterwards. This will limit the waste but allow our users to experiment and try things; yet again, like a good host, we tidy up after our guests and recycle or re-use what we can.

The IT Infrastructure Manager as 'Mein Host'...no more the grumpy chap in the corner but the genial host ensuring that the party runs smoothly yet with a swing. 

Back To Work

After a week away, I come back to the office and find that all is running fine. I have a nice clear day to catch up with emails and things which need attending to. No-one has booked a meeting with me and there is little urgent to deal with. For this, I'd like to thank a number of managers that I worked for in the past who gave me some tips.

1) One manager advised me never to come back to work on a Monday; his reasoning was that if something urgent came up on the preceding Friday, if it was known that you were coming back on Monday; there was a risk that it would be left for you to deal with first thing on Monday whereas if it was thought that you were coming back on Tuesday; it would probably be dealt with. 

I've twisted this, perception often trumps reality; so my 'Out-Of-Office' always says I will be back in the office a day after I am actually back. So generally, I come back to an empty diary and no crises to deal with.

2) Before opening your email, talk to your team and then your boss. You can avoid reading a lot of emails by simply getting them to tell you what you need to know. And if you have managed to keep your diary clear, talking to your team about your holiday can help to extend that feeling of well-being for a little longer.

3) When recruiting and building teams; always make sure that the people that you are recruiting are better than you in at least one thing. This means that you will always have someone to delegate a task to and you can trust them to do an excellent job. Don't be scared of recruiting people who are brighter and more capable than you; as a manager, you are there to lead and not to do everything yourself. 

4) For sanity's sake, have two mobile phone numbers; a work one and a personal one. It is easy to think that combining your work and personal numbers into one and having work pay for your calls as a 'perk'. It's not, it just ensures that you are on call, all of the time for a pittance. Same goes for the Blackberry. 

Having a number that can be turned off which lets you relax and not worry about being interrupted is worth every penny; if you insist, check your work mobile a couple of times a day but if possible, consider turning it off for the duration. If people know you can't be contacted, they'll deal with most situations themselves.

5) Work email; resist the urge to check it. Once people get the idea that you will reply even though you are on holiday, they will abuse it. You can always set up filter rules to forward emails which are really urgent from named subjects/people to a Google Mail or like account.

6) Visit a grave-yard; take note, the world is still turning despite everything!

Oh well, back to work now...