A kind of dns thread management method, device and server based on machine learning

Abstract

Embodiments of the present invention provide a machine learning-based DNS thread management method, device, and server. The method and device are applied to a DNS server built based on a Linux operating system, specifically for acquiring all threads in the DNS server at preset intervals. The number of requests to process requests; use the preset thread control model to calculate the number of requests to get the optimal number of threads; if the current number of threads is less than the optimal number of threads, add a thread; if the current number of threads is greater than the optimal number of threads, reduce a thread. By dynamically adjusting the number of threads of the DNS server, it is possible to avoid insufficient threads during traffic peaks and to avoid waste of computing resources during low traffic periods, thereby improving the efficiency of the DNS server.

Description

technical field

[0001] The invention relates to the technical field of the Internet, in particular to a machine learning-based DNS thread management method, device and server. Background technique

[0002] DNS (Domain Name System, Domain Name System) refers to a set of systems that map domain names to IP addresses on the Internet, enabling users to use domain names that are easier to remember to access the Internet instead of having to remember The IP number string. DNS can be said to be the basis of almost all current online services. In DNS, domain name managers need to build DNS servers to handle DNS query requests. Usually, when the DNS server program is started, a certain number of threads are opened to process query requests in parallel, and each thread is bound to a CPU core to perform related computing tasks.

[0003] However, the inventors of the present application have found in practice that since the query volume of DNS services is closely related to user beha...

Images