|Today (April 28) is last day to register for ESC Silicon Valley. Click here to register. Use promo code Susan30.|
Nobody wants to design an embedded system that fails. But the cost of failure varies greatly across the embedded landscape. A universal TV remote that misses a few commands might be an acceptable product. As electronic systems become more and more autonomous, however, a bug can lead to tragic misbehavior by a car, a chemical plant, or a jumbo jet. Sometimes the cost of error is simply unacceptable.
The dawn of the age of cyber-crime and cyber-warfare has shown an ugly new light on this question of reliability. Yesterday, perhaps, it was sufficient that an embedded system function correctly when operating unmolested and with valid inputs. Tomorrow may demand that the system function safely under attack by ingenious malware.
Reliability in this new world demands rigor at every stage of the development process, from requirements definition through coding and test. This theme runs through many of the conference tracks at Embedded Systems Conference Silicon Valley next week. We have pointed out some of the highlights in the session profiles below, ranging from case studies of disasters to formalizing best practices to reexamining the whole methodology of code development.
If you can attend ESC, please do so. Even if you can’t be there in person, please join in the embedded community’s discussion on this topic, on embedded.com and other forums. This issue really ismission-critical.
Ron Wilson is the editorial director of design publications at UBM Electronics, including EDN, ESD magazine, Embedded.com, the Embedded Systems Conferences, and EE Times' DesignLines. You may reach him at firstname.lastname@example.org.