Caught in the net

John Ballinger, managing director of Adroit Technologies, discusses what implications the introduction of .Net from Microsoft may have on Scada applications.

In June this year Microsoft unveiled its .NET strategy. The fundamental idea behind .NET is that the focus is shifting from individual Web sites to constellations of computers, devices and services that work together to deliver broader, richer solutions.

It is designed to help drive a transformation in the Internet that will see HTML-based presentation augmented by XML-based information. The key difference being that XML-based protocols like Simple Object Access Protocol (SOAP) allow users to interact meaningfully with data instead of just viewing the data in a read-only sense which tends to be the case today.

The situation that currently prevails, with individual Web sites operating as data silos despite large amounts of bandwidth, is similar to the situation common in many companies before the advent of the PC and local area networking, where corporate mainframes and departmental minicomputers were custodians of huge amounts of data.

What does all this mean for Scada – a relatively tiny, if technologically challenging, niche market from an overall IT perspective? Despite its niche market status, Scada has not been immune from the buffeting caused by successive waves of change. First the PC and DOS, then the various Windows generations: Windows 3.x, Windows NT, Windows 9x, Windows 2000.

Each generation has been loosely accompanied by so-called key technologies, including DDE, NetDDE, OLE, COM, ActiveX, SQL, ADO, DCOM, and COM+. Broadly speaking, things have improved for end-user and system integrator alike. Prices have continued to fall in real terms, performance and reliability have generally improved, and the scale of the applications attempted has increased.

Many Scada vendors have attached themselves to the mainstream Microsoft bandwagon over the last few years. This has meant that solutions have tended to be single platform – specifically the Wintel platform. Even COM and DCOM which were meant to herald the era of component software and distributed components respectively, have delivered at best in a limited sense, and then only on the single Wintel platform.

Part of the problem with COM/DCOM is that it provides a tightly coupled solution where objects are instantiated remotely across processes and even machine boundaries. This approach brings complexity and baggage, such as having to maintain object and class versioning information and reference counts. The XML-based SOAP, fundamental to .NET, follows instead a much more loosely coupled approach. This means that computers, devices and services participating in the delivery of some overall end-user solution can, and almost certainly will, run on different platforms: Wintel, Linux, Sun, and so on, provided they all implement SOAP. Not to be left out are WAP capable devices such as PDAs and next-generation mobile phones.

Where does all this leave today’s practising automation engineer or IT consultant, faced with choosing a SCADA, HMI or MES system for an upcoming project? The answer is to go for a system that has at its heart an open, object model, where every property of every object is exposed via a programmable automation interface (Open Automation Technology). A system where features such as clustering and redundancy are fundamental and built-in, not added by user-defined scripts.

A system fully exploiting the currently shipping technologies mentioned above such as ADO, SQL, ActiveX, COM/DCOM. Only in this way can you be reasonably sure that a system you choose today will allow your organisation to fully exploit all the e-Opportunities of the future.

Adroit Technologies

Tel: 01270 627072

On the web