Yikes! Embedded systems.

greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread

I am no embedded systems person but I do feel that first embedded systems and secondly a breakdown of the supply chain are what we have to fear the most. The following is from the JBA web site.

2000.jbaworld.com includes a column from Harlan Smith, a regular year 2000 contributor, and fully supports Harlan's SMCP campaign for global co-operation on year 2000 mitigation and contingency planning.

2000.jbaworld.com is sponsored by JBA International - a leading producer of Year 2000 compliant ERP solutions for the automotive, food, beverages, service management and apparel/footwear industries. Robert Coenen is a Y2K Program Manager at CTA Incorporated in the US - pre-emminent suppliers of embedded systems Y2K services. Here Robert takes us through some of the basic issues of the year 2000 and embedded systems and provides us with his view on the number of embedded systems in existence.

Over to Robert...

The issues related to the Year 2000 and embedded systems are not much different than the issues faced in the Information Technology (IT) arena. Just as there are many lines of code that are at risk in the software arena, there are many programs at risk in the embedded arena. The magnitude of the issue is significantly different.

There are approximately 50 billion chips in the world today. Of those, approximately 5 billion are in computers as we know them, from mainframes down to PCs. The other 45 billion chips are in embedded systems throughout the world. If we assume the same problem magnitude with embedded systems as we have discovered in the software world then between five and seven percent of the embedded systems will have a year 2000 problem. This means between 2.25 billion and 3.15 billion embedded systems will have a year 2000 problem. This number alone is staggering. Beyond the sheer number of potential problem systems is the fact that we don't know where all the embedded systems are, or they are located where we cannot reach them.

Embedded systems can be found in everything from your telephone and VCR through commercial building environmental and energy management and control systems to transatlantic cables and satellites. It is not a simple task to get to a satellite and repair or replace the embedded systems (question? Does the Hubble telescope have any date sensitive systems onboard?).

Add to this the fact that some of these embedded systems are legacy systems for which there is no supporting manufacturer. These systems will require replacement at a potential staggering cost.

Additional issues related to embedded systems, just as with IT systems, are the number of dates that must be supported by the embedded system. It is not just midnight, December 31, 1999 that is a problem. Consider February 29, 2000. This is a special case leap year for date processing. This is the leap year that is divisible by both four and 400, making the Year 2000 a leap year. Embedded systems must correctly transition from February 28, 2000 to February 29, 2000 and then correctly transition to March 1, 2000.

Consider the following. Does the embedded system use dates over time? Does the thermal sensor in the exhaust stack of a coal-fired electrical power generation plant sense the temperature and compare it to previous temperatures? Does this function require a calendar to track the rise and fall of heat in the exhaust stack over time? Will the century date change affect the capability to perform this function? Ask these questions of the many different functions provided by embedded systems. The telecommunications industry must be sensitive to these issues in order to ensure correct billing across the various Y2K dates of interest. So should the banking and other financial institutions that uses date sensitive equipment to calculate interest on monies flowing around the world.

Year 2000 and embedded systems can, and most likely will, have a major impact on the world-wide effort to successfully address and retire the coming century date change. The more technologically advanced the enterprise, whether it be a business or a government, the more likely there will be a year 2000 embedded systems impact.

Robert Coenen Y2K Program Manager

CTA INCORPORATED 900 Heritage Drive, Suite 'A' Ridgecrest, CA 93555 USA

-- Ed (ed@lizzardranch.com), September 29, 1999


Ed....Yourdon? Is this a joke? This old garbage has long been dismissed, please don't restart this nonsense. 1. The numbers are bogus, long dismissed by industry and knowledgeable organizations such as Giga, Gartner Group yes, even this IT group with no epertise in embedded systems (at least in the beginning, lately their reports show that they have obtained better resources in this area) has abandoned the 40 (and now 50?!) billion chip garbage. Experts count systems, not chips, the figures for chips are meaningless drivel. Even HARLAN SMITH laughed at these numbers and the chip counters. Harlan: "To talk about the problem at the chip level is counterproductive and sends people off wasting huge amounts of money looking for "chips in the haystack". Maybe David Hall started this with his 40 Billion chip baloney." See Daily Fix: Embedded Systems Dialog With Harlan Smith

2. "Does the embedded system use dates over time?" Sometimes, but in the vast majority of these cases in minor ways such as date stamping. 3. "Does the thermal sensor in the exhaust stack of a coal-fired electrical power generation plant sense the temperature and compare it to previous temperatures?" No, sensors sense process parameters and convert those parameters to an electronic signal. If a comparison is done at all, it is done by the electronic device that the sensor feeds. You need better examples - and maybe a better expert.

I believe that you will obtain fewer myths and more facts on Y2K in embedded systems from companies who aren't selling Y2K services. For the latest views of Y2K in embedded systems from industry, seeEmbedded Systems Revisited.


-- FactFinder (FactFinder@bzn.com), September 29, 1999.



-- teetime (s@@.com), September 29, 1999.

Yes, this is not helpful at all.

-- Lane Core Jr. (elcore@sgi.net), September 29, 1999.

My Dear Mr. Factfinder

(an odd handle actually) May I be as so bold a to ask you for your credentials. I do not mean the ones you got in collage, nor in some IT shop. I am curious about the ones you would present to companies who build the industrial infrastructure..Not the white shirt nerds in a planning shop. But actual hands on installation and calibration of PLC's and their time sensitive chips...It is this knowledge I am inquiring about. Say for a little back room company like Bechtel Corp. They have put in some rather smallish jobs; I think you would agree. And are not too far down your obivious snob list of prestigeous employers you might wish to work for.

I could go on, But it would be redundant of me. And thank you for your time and consideration


-- Shakey (in_a_bunker@forty.feet), September 30, 1999.

To FactFinder.

I am sure you will be willing to check this address http://www.computerworld.com/home/print.nsf/all/990830BDA6 . In this article you will find that "General Motors Corp. has tested 1.4 million devices at its 150 factories and found that it had to make changes in fewer than 15% of its embedded systems, said John Ahearne, GM's communications manager for the year 2000 project. " Fewer than 15% out of 1.4 million. Isn't it a relief.

-- Boris (MSIS@cyberdude.com), September 30, 1999.

The question begs to be asked. Why sudden silence when Fact Finder is asked for his/her bonifides. Some times silence is not only golden...But reveiling also.


-- Shakey (in_a_bunker@forty.feet), September 30, 1999.

Shakey, Just because someone doesn't answer your questions right away doesn't mean a thing. I drop by here on occassion, but don't always see the posts. In any case, email me and I will be glad to give you some background on my experience. Regards,

-- FactFinder (FactFinder@bzn.com), October 02, 1999.


What about that list of a half dozen verifiable hard failures you said you were going to post...


It might give us an idea of what kinds of systems are at most risk.

-- Linkmeister (link@librarian.edu), October 02, 1999.

Moderation questions? read the FAQ