|
Getting a response
|
22/09/2008
|
| |
Linux poses a dilemma for embedded systems designers. On the one hand, it lets the designer leverage a large pool of developers, a rich legacy of source code and industry standard POSIX APIs. However, on the other, the standard Linux kernel can’t deliver the ‘hard’ realtime capabilities – such as guaranteed response times and microsecond latencies – that many embedded devices require.
The reasons for this are rooted in Linux’s general purpose architecture. Linux was developed to provide UNIX like functionality and has been optimised extensively for the server and IT markets. The requirements of these markets differ dramatically from those of the embedded market and, as such, the tradeoffs chosen in the Linux kernel definition are often inappropriate for embedded applications.
|
| |
Author
Paul Leroux and Romain Saha
|
| |
| |
|
This material is protected by Findlay Publications copyright 2008.
See Terms and Conditions.
One-off usage is permitted but bulk copying is not.
For multiple copies contact the sales team.
|
| |
|
| Email this article |
| |
|
|
|
|