Back office

V-Blocks: it's a new form of packaging for servers, switches and storage for your datacentre, and it looks like the future.

EMC and Cisco have announced that they'll be selling aggregated systems that simply slot into your datacentre. Or 'private cloud', in today's shiny new, updated nomenclature.

Inside the black box will go all the complexity you currently have to manage separately, and the glue to bind them together comes from VMware. It seems like a unique concept, for while Sun tried selling a datacentre in a box and got some traction, this is a different kettle of fish, on a different scale (sorry).

Only HP, with its just-launched Converged Infrastructure concept is anywhere close to this, it seems. The aims of both are similar: to virtualise the hardware and provide a single view of the infrastructure that takes care of the messy details of what VM runs where, what's connected to what, and how. All you need to concern yourself with is the energy bill, security, and general management.

This is the new battleground. Other players are fighting back, including Dell, whose deal with Juniper Networks will allow its resurgent services division to sell combined packages. And IBM has a cloud of products, but it's unclear as yet how well they fit into this new paradigm.

Of course, what this means is that, if you're not in the box, you're locked out. If you're buying, do you care? It's somewhat analogous to the blade server chassis market. There's little commonality and so much vendor lock-in there too, but it's not an issue for most enterprises. And this is a much bigger play which, if successful, promises to bring riches to the pockets of vendors inside the tent.

Will those outside might spend some time getting wet, and find themselves searching for a V-Brolly?

Is email archiving a big problem for SMEs? I spoke recently to Softek, a UK-based disti billing itself as 'highly technical value add distributor of IT security, email archiving and storage solutions'. Naturally enough, it has a view and a product to shift -- I wonder if you'll be totally convinced though.

So we talked about the company's new status as a distributor for PineApp email archiving appliances: boxes that suck in your Exchange (or GroupWise et al) email files after a given period, and delete them after the retention period required for legal compliance has passed. The company makes a series of them, ranging from one said to be suitable for 50 users, right up to 1,000 users.

Softek technical director Mike Bienvenu reckons that the main reason for the existence of appliances such as this, which are aimed at SMEs rather than big enterprises, users are looking for storage management rather than ways of staying legal with respect to email retention.

And that's why, he claims, the PineApp boxes are the only ones of their kind that provide stubbing - the ability to provide a stub for emails that have been archived off, which means the email program -- usually (sadly) Exchange -- sees the email as there. A call for a particular email results in it being pulled off the archive server. Apparently the boxes do deduping too.

I asked why an SME wouldn't just buy more storage rather than go for complex and expensive archiving system. It's not as if they're dealing with massive amounts of data flooding in from all over the globe. The argument for the technology is that compliance means that retaining data for a particular period of time -- up to seven year, for example -- makes this simpler to implement, and that technical issues involving Exchange's storage limitations and the retention of duplicate emails becomes expensive. Bienvenu also said that customers ask for archiving and want to ban PSTs off their network.

I'll confess the story sounded reasonable but not entirely convincing, unless the company's business process involves the generation of terabytes of email data.

Is specialised email archiving technology overkill for an SME?

The world tends to see Ethernet as the way forward for pretty much every type of connectivity requirement. But is it – especially in the context of storage?

This is the debate that once -- but not perhaps so much now -- raged around the idea of Fibre Channel over Ethernet: running FC protocols over the same Ethernet cable as network data. The industry does appear to have accepted that FCoE is the way forward but, for some in the industry, the idea still doesn't make sense.

I'm probably not enough of an expert to call it either way but, during my couple of days at Storage Network World in Frankfurt this week, I've heard some compelling arguments for each position.

Those in favour of the idea of Ethernet everywhere play the familiarity card, citing the success of the technology not just in networking but also in wireless, the wide area network and other places. They also say -- and this is a powerful argument since half the data-carrying cables connect to storage -- that customers are are calling for simplification of the cable plant in their datacentres, which means reducing the number of cables emanating from each rack.

On the other side is the idea that FCoE is essentially SCSI, and that the alterations that Ethernet needs to undergo in order to persuade it to conform to the needs of storage means that, although the cable might be Cat6, what's traversing it isn't Ethernet. And that anyway, it doesn't matter whether it's Ethernet or not in this context.

Behind the naysayers' words is also a belief that one large -- make that very large -- networking vendor dreamed up the idea of FCoE in order to create a bottleneck in the standards committees because its own technology simply wasn't ready at the point that storage vendors were ready to move FC forward.

Far be for me to accuse Cisco (oops) of sandbagging: I'm sure the company is far too honourable to play so underhand a trick. But could it be that this is actually what happened? Is FCoE really the answer to the need for a single cable? And does it matter now anyway?

What do you think?

Virtualisation has moved from being a shiny new technology concept to a sticky business that involves getting your hands dirty with some real technology and business problems. But while server virtualisation is now a well-known technique for saving money by increasing server CPU utilisation, storage virtualisation remains a hazy concept.

It means you can use the term how you like. Perhaps the best-known use of this technology is tape virtualisation, where your automated library of tapes takes care of exactly where your backups are stored, and just returns the data you want. Eventually. In theory.

On the other hand, virtualising rotating media is less well-known, mainly because it means different things to different people. Here at Storage network World in Frankfurt, the most – er – innovative, but perhaps instructive, use of the term so far was by analyst Tony Lock, who pointed out that every file system is a form of storage virtualisation.

How so? At heart, virtualisation is about hiding the underlying hardware from the software and management layers above, allowing the technology to take care of the intricate details of where an application is run, or data is stored. Given that definition, a file system qualifies.

So even when I was running a BBC Micro back in 1982, it included storage virtualisation. Who knew?

So when the vendor sidles up to you and whispers seductively about storage virtualisation, make sure you're both talking about the same thing. I'd be interested to learn what storage virtualisation means to you. No, really...

Are flywheels better than batteries? I visited a datacentre a while back that had a flywheel to fill the gap between power outages and the diesel generator kicking in. The datacentre manager told me it made sense because batteries - the usual alternative - were expensive to maintain and to cool

"Batteries would need a chiller full time to keep them cool, which means 250kW off the top of our power usage, compromising our density proposition. Also, batteries aren't particularly green – although green is important, most customers won't pay extra for it," I was told.

Now Chloride, a battery maker - oops, a company that used to make batteries but now focuses on UPSes - tells me that this is hogwash and that if the datacentre manager had come to them he'd have found that flywheels have their own issues (of course) and that batteries hardly heat at all as they only do so when working - charging or discharging - and that most of the time they're doing neither.

Which is right? What's your take on batteries versus flywheels? (That's big lumps of rotating metal, not Wolf J. - look it up!)