Multicast Packet Replication Table Structure with Mode Field
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing replication list table structures for multicast packet replication suffer from poor usage efficiency and limitations in entry-sharing, leading to suboptimal performance in managing packets with different headers across multiple output ports.
Innovation Solution
A new replication list table structure is introduced, featuring entries with four fields: End, Mode, REP_ID0, and REP_ID1, where Mode determines how to modify packet headers, and REP_ID fields can be used for jumping or calling other entries, allowing for flexible configuration and reduced reliance on Next_ptr fields, enabling efficient packet replication and entry-sharing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the existing RPL table structure with REP_ID and Next_ptr fields is used, then the switch can read entries one by one to realize packet replication, but the usage efficiency is poor due to consuming more fields
Solution Approach 1:
The patent combines multiple REP_ID fields (REP_ID0 and REP_ID1) into a single entry, allowing two replication operations to be represented in one entry instead of requiring separate entries. This merging reduces the total number of entries needed in the RPL table, thereby improving usage efficiency while reducing the complexity of traversing through multiple Next_ptr fields.
Solution Approach 2:
The patent introduces a Mode field that enables a single REP_ID1 field to serve multiple functions: it can either indicate another REP_ID for header modification (first type) or indicate the next entry to jump to in the list (second type). This multi-functionality eliminates the need for separate Next_ptr fields, improving space utilization and usage efficiency while maintaining the ability to traverse the table structure.
2Productivity
If multiple REP_ID fields are added to reduce Next_ptr consumption, then usage efficiency improves, but more REP_ID fields are consumed for certain output ports
Solution Approach 1:
The Mode field enables REP_ID1 to be universally applicable for both header modification indication and next entry jump indication. This multi-functionality means that the same field can serve different purposes depending on the Mode setting, reducing the overall quantity of fields needed in the table structure while maintaining flexibility for different replication scenarios.
Solution Approach 2:
The patent changes the parameter interpretation of REP_ID1 based on the Mode field value. When Mode indicates the first type, REP_ID1 is interpreted as a header modification identifier; when Mode indicates the second type, REP_ID1 is interpreted as a jump target. This parameter change allows the same field to convey different information, reducing the quantity of fields required while improving usage efficiency.
3Productivity
If entry-sharing is implemented in existing structures, then usage efficiency improves, but sharing is limited to end positions only
Solution Approach 1:
The Mode field provides universal functionality that enables entry sharing at any position in the table, not just at the end. By using REP_ID1 as a jump target when Mode indicates the second type, any entry can share replication information with any other entry in the table, providing flexible entry sharing throughout the entire table structure rather than being constrained to end positions only.
Solution Approach 2:
The Mode field acts as an intermediary that mediates between different entry types and enables flexible sharing. It allows the system to switch between interpreting REP_ID1 as a header modification identifier or as a jump target, thereby enabling entries to share replication information dynamically based on the Mode setting, increasing adaptability and versatility of entry sharing.
Data Source
AI summary
A replication list table structure for multicast packet replication is provided. The replication list table structure includes a plurality of entries. Each one of the plurality of entries includes a first field, a second field, a third field and a fourth field. For each one of the plurality of entries, the first field is used to declare whether the entry is an end of a program execution, the second field is used to declare the fourth field as a first type field for indicating a switch how to modify a header of a multicast packet, or as a second type field for indicating the switch, while reading the list, to jump to another one of the plurality entries, and the third field is preset to the first type field for indicating the switch how to modify the header of the multicast packet.