Is one application powerful enough to be the core lever on which any enterprise can make a significant shift in infrastructure? And if, in theory, such a thing is possible is it likely that application will be Customer Relationship Management (CRM)?

The question comes to mind because of last week’s announcement of a market report from Research and Markets highlighting the impending rush of companies like Oracle, Microsoft and SAP towards the SaaS (Software as a Service) marketplace. CRM is the market the analysts think these major players will be pitching at initially, and on the face of it there could be a certain sense of safety to such a move – but only on the face of it.

It is, after all, the market where the adolescent SaaS industry’s `poster child’ company, Saleforce.com, has constructed its reputation. But even Salesforce has sensed it probably has a much better business proposition if it can offer users more than the one application. For example last year, just before it added some ERP and Content Management capabilities, it turned in a loss. It has also made much, recently, of its AppExchange tools which allow users some degree of freedom in creating their own applications.

Another SaaS specialist, NetSuite, has been offering similar development capabilities for over a year, and could be of worth watching, if only because it is heading for its IPO on the US stock markets and has Oracle’s CEO, Larry Ellison, as one of its significant private investors. Netsuite - as its original company name, NetLedger, implies – found its niche in providing accounting and business management services, an arguably wider `niche’ than CRM alone.

That is the type of sector that would map more closely onto Oracle’s core marketplace than CRM alone. SAP could certainly target the wider realms of ERP than just sticking to CRM. It is certain that these wider markets are where those two vendors are heading, but is CRM a good starting point?

Any business with a traditionally installed and owned infrastructure, particular if it is of significant size, is going to look suspiciously at SaaS as an alternative option. But despite the many provisos, qualifications and doubts that senior IT and business management might set against such a move, the arguments put forward in its favour – better economics, manageability, flexibility and business agility – mean it has to be considered as a real contender for future infrastructure provisioning. So starting a piecemeal implementation approach now is sensible from the perspective of both enterprise users with large infrastructures, and vendors with complex applications environments.

But I doubt that either party is likely to learn enough, quickly enough, from going down the CRM route. If a piecemeal approach is only picking off small, departmental systems as `experiments’ they might just opt to go with the specialists who already know what they are doing and would be very happy to treat an enterprise department as a small or medium sized business unit, especially if they have a long track record of SaaS-delivered CRM.

The likes of Oracle, SAP and Microsoft have many more applications opportunities than CRM available, and they would learn far more from selecting some of them. There would also be some real grounds for competition in the market – for example, I’m sure Microsoft would love to squeeze a SaaS-delivered SQLServer application or two into a rock-solid Oracle shop, and the technology is ideal to allow that to happen.

But such big names all just going for CRM as their starting application smacks of timidity at least, though I suspect they will call it `circumspection’.

Let’s face it, virtualisation technologies are a very good idea, particularly from the point of view of getting far better utilisation of the IT infrastructure. From the infrastructure management point of view, that utilisation is just so damned flexible – it can be used to cut long-term Total Cost of Ownership simply by making sure individual servers work harder to earn their keep, or it can be used to rapidly and flexibly scale business processes to meet fluctuating demands in levels of business.

But it can bring with it some downsides of increasing operational complexity and potential security holes as servers find themselves changing their `personalities’ and workloads. Managing that is an issue many businesses will face as they start to exploit virtualisation, but it is fair to guess that not many actually realise that such management capabilities are necessary.

But without them there is a growing potential for poorly managed virtualised infrastructures to gradually become a dog’s breakfast, a mess of tasks that may have completed from a business point of view, but have not been managed – well – tidily. Defragging a PC disk to tidy it up can be a pain – imagine the idea of `defragging’ a complex virtualised infrastructure.

Tools are now starting to appear that address this issue, such as Configuresoft’s recently introduced Configuration Intelligence tool, aimed at providing at least some degree of management over VMware installations. This tool raises an interesting debating point about this whole area. Yes, the likes of IBM, HP and BMC have the comprehensive, big systems solutions for these problems that cover most, if not all of the configuration and security management issues. But for many users this level of solution is going to be not just overkill, but expensive overkill.

There have been criticisms levelled at the Configuresoft solution, mainly pitched at its limitations. The tool is aimed squarely at managing the virtual machine component of virtualisation, and does not really cover the next level of abstraction of the technology – the way in which it can be used to improve overall infrastructure performance or scalability, or enhance business agility. And arguably, it is in these areas that poor management that will have the biggest negative impact on the majority of virtualised environments.

Managing virtual machines – being able to manage the individual configurations of a host server and its guest virtual servers - is most certainly an important capability but virtualisation, even for an SMB-sized installation, is already a wider subject than that.

What this shows is that there is now a real user need to manage virtualised infrastructures of all sizes, and there are high-end, comprehensive (and in reality none-too-cheap) solutions and cost-effective but limited solutions lower down the scale. Both are good, but they don’t necessarily fill the hole that now exists between them.

Not to put too fine a point on it, that represents a big market opportunity for some company or other.

At the back end of last year Sun Microsystems came up with Project Blackbox, a good marketing wheeze in the form of a `datacentre in a box’. The idea is interesting but, given the nature of the box in question, it seems to be a little too specialist for it to be widely successful. In practice, as is often the case, there is a clear division between an idea and the specifics of an implementation of it.

For example, the `datacentre in a box’ idea has emerged again, but this time in a much smaller form and aimed at the small to medium-sized business community. What has that got to do with enterprise infrastructures? Well, quite possibly a good deal, at least conceptually.

Let’s go back to Sun’s box to make the point. This is due to appear around now and is some box. Yes, it can hold 250 SunFire servers with up to 2 Petabytes of storage. But it comes in a 20 x 8 x 8 foot shipping container which, although Sun argues is an economic form factor, is still not cheap. Not only that, the logistics of physically locating such a beast and feeding it with power and network connections will, I suspect, prove to be as logistically problematic for most enterprises as trying to build a similar datacentre inside an office block.

They could perhaps find uses providing the onsite resources needed in major civil engineering works, stacked amongst the Portacabin offices, but there has to be a risk of someone coming along with a big crane, a low-loader and a several sets of wire cutters…..bye bye box. All in all, it seems to suffer from sledgehammer/nut syndrome when looked at in the cold light of day.

And there the idea may have stayed, except that an Indian company HCL Infosystems, has come up with a much smaller implementation of the idea, aimed at the SMB sector. It is based on x86-architecture servers running either Windows or Red Hat Enterprise Linux, and the company says it has partnerships set up with ISVs to provide typical applications such as ERP, CRM, mail management and the like. So far so OK, but the boxes can also be clustered, the company has come up with its own solution for this task.

Clustered boxed datacentres based on industry-standard technology give enterprises the potential to build infrastructures quickly and flexibly – and even keep them indoors. It is an approach that is full of compromises, but that is what makes the building blocks of so many workable solutions. For example, it could allow the creation of a semi-distributed datacentre, where boxes offer sufficient performance to provide physical departmental server consolidation, while being integral to a logical single datacentre via high bandwidth networking. There is a compromise between energy consumption, centralised management, logistical flexibility and the provision of end user services.

It could also make a very good `quick and dirty’ compromise for scaling out services in a packed and logistically entrapped datacentre.

I suspect we may see the likes of HP and IBM moving to market their SMB-targeted Blade chassis systems along similar lines over the coming months as they realise that many enterprises, no matter what they might like to do with their IT infrastructures, come to learn that quick and dirty compromises are likely to be their best pragmatic solution.

The trouble with High Performance Computers (HPC), otherwise known as supercomputers to the hoi poloi, was neatly summed up at the recent International Supercomputer Conference in Dresden by Burton Smith, Microsoft’s recent `acquisition’ as Technical Fellow, from jobs such as Chief Scientist at supercomputing specialist, Cray. “HPC applications are the ones that run on HPC systems, and an HPC system is a machine that runs HPC applications.”

And it’s true: HPC is the domain of propellor-headed rocket scientists who have little in common with the real world of mainstream IT. There is nothing that that is relevant to IT infrastructures.

If you stop the clock now, then that is true; but if you let it run on only a little bit it becomes remarkably easy to see that there is in fact a great deal mainstream big systems IT managers not only can learn from HPC. Not only that, but they will be left in a hole, digging furiously, if they don’t learn it.

The key thing to learn from HPC is that the majority of the Top500 supercomputers use multicore x86 architecture processors – and so do the majority of mainstream servers in IT. Yet in the mainstream world we are still stuck with single-threaded applications as the norm, while in HPC they long ago cracked the parallel processing problem – that’s where the performance comes from.

This could remain a chalk-and-cheese differentiation if not for the fact that Moore’s Law makes a simple prediction. We will have 8-core, x86 processors by the end of this year, so by 2012 we will have 128-core, x86 architecture processors. In other architectures, there will be processors with over 100-cores by the end of this year: Cisco’s Metro device is pitching at 188-cores. So predictions of over 1,000-cores per socket and 1 million-cores per system by 2013 seem entirely reasonable.

Put simply, in five to six years time, the single-threaded, single program counter architecture that has been the norm of mainstream computing for half a century or more has to fade away. Without that, we will be stuck with systems infrastructures that won’t go any faster, will do no more work, no matter how much is thrown at them in the way of hardware resources.

To exploit multicore processors, the only sensible option now is to look to what has been learned about parallel architectures in the HPC world and adapt it for the mainstream. Infrastructures based on parallel architectures will certainly map well onto coming service-delivery architectures now coming along, for example. It could also be the real lever to giving Software as a Service (SaaS) providers with platforms that can deliver both real depth and breadth in their service offerings – to the point where many users may question whether, or to what extent, they require their own infrastructure at all.