Multi-level processing method for temperature alarm of heterogeneous many-core processors

Abstract

The invention discloses a multi-stage processing method for temperature alarm of a heterogeneous many-core processor, which comprises the following steps: a processor monitoring system detects temperature information of the processor in real time through a temperature sensor, and transmits the detected temperature information to the monitoring system ; When the value of the temperature information received by the monitoring system is higher than the set temperature threshold, an alarm message will be issued, and the frequency and voltage reduction operation will be performed on the main core and slave core of the processor node where the warning occurs. Track and record the temperature information of the corresponding processor node during the pressure operation; compare the temperature information of the processor node with the set temperature threshold within the set time interval, if the value of the temperature information of the processor node is lower than the set temperature threshold, restore the processor to a normal operating state, otherwise, go to the next step. The present invention adopts a multi-level processing method to avoid prematurely adopting a large-cost control method, and takes into account both effectiveness and cost.

Description

technical field [0001] The invention relates to a multi-stage processing method for temperature alarms of heterogeneous many-core processors, belonging to the technical field of computers. Background technique [0002] A heterogeneous many-core processor is a high-performance heterogeneous central processor that integrates a small number of general-purpose master cores for management, communication, and computing functions and a large number of streamlined slave cores for computing functions on a complete chip. The general-purpose main core runs a general-purpose operating system, mainly responsible for the management and control functions of the entire chip, and also undertakes certain computing functions and communication functions between the chip and the outside. From the core to the function of accelerating calculation. [0003] Abnormal temperature of the processor can easily make the processor run unstable and cause the processor to malfunction. In a large-scale com...

AI Technical Summary

Problems solved by technology

[0004] At present, the temperature abnormality of a common single processor is generally dealt with by reducing the frequency and voltage. In a large-scale system, in addition to the processing means of a single processor, there is no systematic temperature control method. For larger-scale operations, it is generally through Recovery is reserved for possible failures, there is no special means of ensuring continued smooth operation of a processor node that may fail due to temperature anomalies

Images