A message forwarding method and device
By reading the token granularity template from an independent register in the network processor, generating tokens, and forwarding packets, the problem of DDR bandwidth and DRAM memory overhead caused by increasing the expected rate is solved, enabling flexible traffic shaping and performance optimization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NEW H3C SEMICON TECH CO LTD
- Filing Date
- 2025-06-23
- Publication Date
- 2026-07-21
AI Technical Summary
Existing technologies, while increasing the expected speed of network processors, lead to increased DDR bandwidth overhead and DRAM memory overhead, thus affecting the performance of network processors.
By reading parameters from the token granularity field and the representation granularity identifier from the independent register, the target token granularity template is determined, a token is generated, and the message is forwarded according to the amount of forwardable data. This avoids directly determining the token granularity based on the value of the token granularity field in the QD and utilizes the independent register to store more token granularity templates.
It achieves the goal of increasing expected speeds while reducing the impact on network processor performance, and flexibly responds to complex and diverse business scenarios.
Smart Images

Figure CN120512403B_ABST