Messages in this thread | | | Date | Tue, 13 Aug 2002 23:54:48 -0500 | From | watermodem <> | Subject | Symptoms of overload on Southbridge <-> Northbridge |
| |
This question is not truly a kernel question but it has kernel attributes so I decided to ask it here.
I have some measurement situations (using P-3 and P-4) machines where the Southbridge to Northbridge connection could get saturated and maybe even hit the bandwidth limits. These machines are usually older uni-processor Dells with one flavor or another of Intel chipset.
What sort of hardware symptoms can I expect to see when the bus overloads?
(I am not that worried if the OS crashes but am worried about possible hardware damage - heat, out of sequence drive commands etc...)
The machines will have 1 gigabit ethernet taking a full sniffer port load from a loaded CISCO 6509 switch and many more 100Mbit ethernets, several sniffing and the rest communicating. (Adaptec 4 port units) I know that is an insane load but we have hundreds of labs many being near idle most of the time. The mainly near idle labs could use the cheaper equipment. It would be nice if there was a way to sense nearing overload and dynamically "throttle down". Would any of the /proc provided statistics give me the clues I am looking for?
Thanks...
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |