How do you design a computing system to provide continuous service and to ensure that any failures interrupting service do not result in customer safety issues or loss of customers due to dissatisfaction? Historically, system architects have taken two approaches to answer this question: building highly reliable, fail-safe systems with low probability of failure, or building mostly reliable systems with quick automated recovery. The RAS (Reliability, Availability, Serviceability) concept for system design integrates concepts of design for reliability and for availability along with methods to quickly service systems that can't be recovered automatically. This approach is fundamental to systems where the concern is quality of service, customer retention, and due diligence for customer safety.
From the end-user standpoint, computing systems provide services, and any outage in that service can mean lost revenue, lost customers due to dissatisfaction, and, in extreme cases, loss of life and possible legal repercussions. For example, with cell phone services, if it's only in rare cases that I can't get a signal or make a connection, I'll stick with my service provider. But if this occurs too often or in critical locations like my office or home, I'll most likely switch providers. The end result is loss of revenue and loss of a customer. Embedded systems not only provide value added services such as communications, but they also provide critical services for human safety. For example, my anti-lock braking system is provided by a digital control service activated by my ignition. My expectation is that this service will work without failure once ignition is completed. Any system fault that might interrupt service should prevent me from using the vehicle before I start to drive. A failure during operation could result in loss of life and product liability issues.
Read this full article at ibm.com
|
|
Only registered users can write comments.
Please login or register.
Powered by AkoComment!