Data processing method and device

Abstract

The invention discloses a data processing method. The method includes: traversing the rules in the aging time linked list according to a predetermined traversal mode, and marking the rules in the aging time linked list whose establishment time is greater than the predetermined time threshold during the traversal process is invalid, wherein, the predetermined traversal method is to traverse according to the length of the establishment time from short to long; the rules in the hash table corresponding to the rules marked as invalid in the aging time linked list are also marked as invalid; the hash table During the traversal process, the received data headers are rule-matched and the rules marked as invalid in the hash table are deleted; the rules corresponding to the rules deleted in the hash table are deleted in the aging time list. The invention can avoid data reading and writing conflicts between the rule matching task and the rule aging task, realize parallel processing of the rule matching task and the rule aging task without locking, reduce password overhead, and improve system performance.

Description

technical field [0001] The present invention relates to the field of message processing, in particular to a data processing method and device. Background technique [0002] In the field of network security, network security filtering equipment needs to analyze each data packet on the network, so as to realize the transparent transmission of normal traffic, the forwarding of suspicious traffic, and the discarding of illegal traffic. In order to realize the above processing of traffic, it is necessary to These flows are classified, and the classification process is the matching process of data (that is, packets) and rules. [0003] Such as figure 1 As shown, when filtering traffic, generally three functional modules are required, namely the protocol identification module, the rule matching module and the data processing module, so as to realize transparent transmission of normal traffic, forwarding of suspicious traffic, and discarding of illegal traffic. Among them, When id...

AI Technical Summary

Problems solved by technology

[0006] It can be seen from the above specific description that for the existing technology, the solution to resolve rule read-write conflicts by locking the hash flow table of the operation is essentially a task serialization process, so in the present In some computer environments, especially on multi-core systems, forced serialization will inevitably lead to a decrease in system performance, and the process of locking often means an increase in password overhead. Therefore, whether it is a lock implemented by the kernel or implemented in user mode All locks are a very time-consuming process. In addition, since the processing time for rule aging is very short (that is, at the millisecond level), it is obvious that the process of adding and unlocking is very frequent, which will obviously reduce system performance, and even degrade system performance to an unacceptable level

[0007] For the above-mentioned problems in related technologies, no effective solution has been proposed yet

Images