Operating computer memory

摘要

Operating computer memory in a computer including dynamically monitoring, by a predictive failure analysis (‘PFA’) module, correctable memory errors and memory temperature and managing cooling resources in the computer in dependence upon the correctable memory errors and memory temperature.

主权项

1. A method of operating computer memory in a computer, the method comprising:
dynamically monitoring, by a predictive failure analysis (‘PFA’) module, correctable memory errors and memory temperature; and managing cooling resources in the computer in dependence upon the correctable memory errors and memory temperature, wherein managing cooling resources includes controlling a fan speed; including:
determining, upon identification of a current correctable memory error from historical error data, a current correctable memory error rate; andoperating the computer memory with additional cooling when the current correctable memory error rate is between a dynamically adjustable baseline error rate and a maximum error threshold rate and when the current memory temperature is greater than a predetermined threshold temperature, including:
determining whether the current correctable memory error rate is greater than the maximum error threshold rate;if the current correctable memory error rate is greater than the maximum error threshold rate, logging a PFA alert;if the current correctable memory error rate is not greater than the maximum error threshold rate, determining whether the current memory temperature is greater than the predetermined threshold temperature,if the current memory temperature is greater than the predetermined threshold temperature, determining whether the current correct memory error rate is greater than the baseline error rate;if the current correctable memory error rate is greater than the baseline error rate, increasing cooling in the computer and updating the historical error data in accordance with the identified current correctable memory error; andif the current correctable memory error rate is not greater than the baseline error rate, operating the computer memory without increasing cooling and updating the historical error data in accordance with the identified current correctable memory error.