An integrator’s paradise?

So what’s going to become of the SCADA suppliers – and the systems integrators that are currently doing so well out of engineering their software? It’s a serious question. As Sun, Microsoft and so on continue to push the boundaries of thin-client/server, Web/network-orientated computing, it’s looking ever more likely that the Internet/intranets will provide for […]

So what’s going to become of the SCADA suppliers – and the systems integrators that are currently doing so well out of engineering their software? It’s a serious question. As Sun, Microsoft and so on continue to push the boundaries of thin-client/server, Web/network-orientated computing, it’s looking ever more likely that the Internet/intranets will provide for generic information viewing.

So, will the SCADA companies no longer have a rOle? And, as connectivity ceases to be an issue, will the rich pickings for systems integrators also vanish?

Perversely perhaps, the SCADA package vendors are rushing to embrace the new technology. They’re falling over themselves to cry `Java’, `ActiveX’! Fact is, they have little choice. Just as the DCS vendors before them found that they had to `open’ up or be left behind, so with the SCADA suppliers.

Yet it seems the tighter their embrace, the faster looms their demise. It’s being hastened also by the growth of ActiveX Controls and JavaBeans components. Surely soon every function you care to imagine will be a component sitting in a library just waiting to be dragged-and-dropped into your application? And if so, who needs SCADA packages? You can build your own!

Two other aspects come to mind. One concerns the nature (chip hardware, operating system footprint, etc) of the beasts at the distributed client end. Processors are getting still smaller and more powerful. Sys.Build, this month carries news from Motorola, Siemens, Microsoft, Sun, Dec and Hewlett-Packard that already has a bearing. The next generation of microcontrollers, ROMable Windows NT from VenturCom and now Windows CE will drive change.

Second, consider network architecture. Already, the old pyramid model of layered networks is just that – old! Few consider the real world to work like that. The path from sensor and field I/O through PLCs, SCADA, MIS and CIM is already a jumbled one. The ideal is peer-to-peer producer-consumer (see page 16, Sys.Build) – a flattened architecture providing for all data sources.

And, although there are as yet physical limits for copper wire in terms of bandwidth, the digital world is not standing still here either. At most, there’s only room for say three network types now. And how many protocols? Look at what’s happening with software bridges already – OPC-style.

So what will client-server look like in the near future? Will even the `lowest level’ go Browser-style – not on the Web, but with similar look, feel and flexibility? Will our transmitter/controllers be able to dip into others’ data and servers’ databases and applications – via a Browser-ette?

Seems we’ve gone full circle. For systems integrators, the sheer spread of new possibilities will keep them going. They thrive when change, big manuals and complexity abound! And as for the SCADA companies, they’ll be continuing to re-invent themselves, surfing on with the latest and greatest ideas – whoever develops them.

BRIAN TINHAM