At the beginning of the year I published a set of technology predictions for 2005. Now, just after the half year point, would probably be a good time to review them. So here we go...

Linux Server Growth

The original prediction suggested that Linux would:

"continue to grow its server market share. The "commodity OS" market will be driven both by the applications that are available (on Linux and Windows) as well as the perception of value for money. For this reason, we don't expect Windows market share to diminish greatly – as it still provides the richest application environment by a considerable margin."

Market figures published by IDC in May put unit shipment growth in the server market at 13.5% with Linux growth at 31.3%, Windows growth at 10.7% and Unix growth at 5%. From these figures, the trend is clear. Linux is slowly crushing Windows and Unix.

It is a little more complex than that, of course. What is happening is that Linux is gaining credibility in areas of server usage where it was weak; package implementations, database apps, decision support and so on. In some areas such as the implementation of Microsoft packages (Exchange, SharePoint, Microsoft Business Solutions, etc), Linux may never get a grip and Windows growth will depend entirely on the popularity of the packages.

One thing to note about these figures is that they will not take into account two factors which increase the number of Linux instances:

1. The repurposing of old Intel based hardware
2. The creation of Linux partitions on mainframes

I have been unable to find an accurate source of figures for this use of Linux so it is difficult to say how much one would need to adjust IDC's figures to account for these additional Linux factors. Not that it matters much – the trend is clear.

Nothing has occurred that would lead me to alter my opinion of the continuing validity of this trend.

Utility Computing

I wrote a long introduction in the original article that I don't intend to reproduce here. The gist of the prediction was that we would see further acquisitions by the major "Utility Computing" players in order to fill out their system management portfolios.

"In 2005 we expect to see a great deal of evolution in this area. There will undoubtedly be further acquisitions aimed at strengthening each vendor's position (by IBM, Hewlett-Packard, Sun Microsystems, Microsoft, Oracle, EMC, Computer Associates and BMC – all of whom count themselves as participants in this market). We regard system management as a key element of delivering utility computing."

First of all, this prediction has been reasonably well fulfilled so far. Oracle acquired Oblix. CA acquired Niku and Concord. BMC acquired OpenNetwork and Calendra. IBM acquired Isogon (for asset management) and it also acquired SRD – which has application in identity management but also elsewhere. Sun's acquired StorageTek, giving it a good deal more than some useful ILM software products. Sun's acquisition of See Beyond is, in my view also related to this market battle – even if it has more to do with SOA.

We should expect to see more acquisitions from these vendors throughout this year. It is still the case that none of these vendors has filled out its "infrastructure management" portfolio and most will choose to buy rather than to build.

IT Security

In this area I noted that the greatest IT security weaknesses for most enterprises are:

1. The lack of an integrated set of security products to manage the IT security problem centrally and coherently.
2. Inadequate identity management technology and authentication deployed.

This is still the case and it will remain so for quite a while. Coherent IT security takes a long time to build, especially in large organizations, and right now none of the major vendors can claim to offer a full solution that covers both points described above.

I predicted that:

"There will be a great deal of merger and acquisition activity as well-positioned vendors (IBM, Cisco, Hewlett-Packard, Sun Microsystems, CA, BMC, Novell, Symantec and others) attempt to broaden their security portfolios and services. We will see some vendors marketing "comprehensive security platforms" in anticipation of being able to deliver on such a vision. In the meantime, IT Security will still prove to be a major headache for all organisations. We expect 2005 to be every bit as bad as 2004 was."

2005 has so far been every bit as bad as 2004 for security breaches – at least it seems that way with the high profile hacks of ChoicePoint, Lexis Nexis and CardSystems Solutions all coming to light. However it is probably not as bad as it seems.

IT security acquisitions are happening, of course (many of those mentioned under system management above are for identity management, a technology which spans both IT security and system management): Also Symantec acquired Brightmail and Novell acquired Immunix. There are also IT security acquisitions happening lower down in the market. For example, Consul acquired Netmon2.

IT security is such a diverse area (a $13 billion market) and there have been so many start-ups in this space (literally thousands), that consolidation is not so much a prediction as an inevitability. We can expect the trend to continue not only for the rest of the year, but for a few years yet.

If you enjoyed this article, why not read the next three reviews of my technology predictions for 2005...