PCI-Express Replay Buffer Indexing for Null Packet Skip
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network systems face challenges in efficiently identifying and resending packets from a retry buffer, particularly in excluding nullified packets to avoid wasting bandwidth and preventing confusion for downstream receivers.
Innovation Solution
A packet outputting device with a retry buffer and an index table that uses unique sequence numbers to track and locate resendable packets, allowing for skipping over nullified packets by shifting entries in the index table and using a subset of sequence number bits to generate an index for fetching the correct packets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all packets in the retry buffer are resent during mass replay, then packet delivery reliability is improved, but bandwidth is wasted on nullified packets and downstream receivers experience confusion
Solution Approach 1:
The patent extracts and removes nullified packets from the retry buffer before mass replay operations. The system identifies packets marked as nullified (due to errors or retransmission) and excludes them from the replay sequence, thereby preventing bandwidth waste and receiver confusion while maintaining reliability for valid packets
Solution Approach 2:
The system uses feedback from packet transmission status and error detection mechanisms to identify which packets are nullified. This feedback information is used to dynamically adjust the mass replay operation, skipping nullified packets and only retransmitting valid packets that need recovery
2Measurement precision
If packet sequence numbers are used to locate packets in the retry buffer, then packet identification accuracy is improved, but system complexity increases due to index table management
Solution Approach 1:
The patent implements preliminary organization of packets in the retry buffer using sequence numbers as indices. An index table is pre-established to map sequence numbers to packet locations, enabling efficient O(1) packet retrieval during mass replay operations without complex search algorithms
Solution Approach 2:
The index table serves as an intermediary data structure that simplifies the relationship between sequence numbers and packet locations. Instead of directly searching through the retry buffer, the system uses the index table as a mediator to quickly locate packets, reducing the complexity of packet identification
3Reliability
If the retry buffer stores all egressing packets for potential replay, then packet recovery capability is improved, but memory usage increases
Solution Approach 1:
The patent extracts and removes nullified packets from the retry buffer to free up memory space. By identifying and eliminating packets that are marked for nullification (due to errors or successful retransmission), the system maintains packet recovery capability for valid packets while reducing overall memory consumption
Solution Approach 2:
The system discards nullified packets from the retry buffer after they have been processed or determined to be unrecoverable. This selective discarding approach allows the system to maintain necessary packet recovery capabilities while optimizing memory usage by removing packets that no longer contribute to reliability
Data Source
AI summary
In PCI-Express and alike network systems, back-up copies of recently sent packets are kept in a replay buffer for resending if the original packet is not well received by an intended destination device. A method for locating the back-up copy in the retry buffer comprises applying a less significant portion of the sequence number of a to-be-retrieved back-up copy to an index table to obtain a start address or other locater indicating where in the retry buffer the to-be-retrieved back-up copy resides. A method for skipping replay of late nullified packets includes deleting from the index table, references to late nullified packets.