PCIe Dual-Plane Switching for Reliable Packet Transmission
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The current Peripheral Component Interconnect Express (PCIe) switching network's tree-structure architecture limits system reliability due to its point-to-point communication mode, leading to bottlenecks and reduced reliability when failures occur, especially in high-bandwidth concurrent reading and writing operations.
Innovation Solution
Implementing a dual-plane switching network with a PCIe interface controller that manages packet transmission between active and standby switching units, using a reliable transaction layer packet (TLP) transmission module to ensure that packets are sent through both paths simultaneously, with selective receipt based on source ID and sequence number, and includes features like TLP transmission queues and acknowledgment mechanisms to maintain sequence and reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a tree-structure PCIE switching network is used, then system simplicity and ease of manufacture are improved, but system reliability deteriorates due to single-point failures affecting all CPUs
Solution Approach 1:
The network is segmented into multiple independent planes (active and standby), where each plane operates separately. This segmentation allows failure isolation within individual planes while maintaining overall system functionality through redundant planes, resolving the contradiction between structural simplicity and reliability.
Solution Approach 2:
The network topology parameter is changed from a single tree structure to a multi-plane structure with active and standby configurations. This parameter change enables the system to maintain simplicity in control logic while improving reliability through redundant pathways that can be activated upon failure detection.
2Device complexity
If a single PCIE switching unit is used, then device complexity is reduced, but reliability deteriorates due to single points of failure
Solution Approach 1:
The system implements beforehand cushioning by pre-configuring standby PCIE switching units that remain inactive during normal operation but are ready to take over immediately upon failure detection. This approach maintains low device complexity during normal operation while providing high reliability through pre-prepared redundancy.
Solution Approach 2:
The standby switching units are created as copies of the active unit, replicating the necessary functionality and configuration. This copying approach allows the system to maintain simplicity by using identical modular units while improving reliability through redundant copies that can replace failed units.
3Reliability
If packets are sent through multiple paths simultaneously, then reliability is improved through redundancy, but device complexity increases due to dual-plane management
Solution Approach 1:
The system uses periodic action through sequential switching between active and standby planes based on failure detection events. Rather than continuously managing both planes simultaneously, the system switches planes periodically or event-driven, maintaining reliability through redundancy while reducing complexity by actively managing only one plane at a time under normal conditions.
Solution Approach 2:
The PCIE interface controller acts as an intermediary that manages the dual-plane configuration, handling the complexity of redundant path management while presenting a simplified interface to upper layers. This intermediary absorbs the management complexity, allowing reliable multi-path transmission without exposing the full complexity to the rest of the system.
4Reliability
If acknowledgment mechanisms and sequence numbers are implemented, then transmission reliability is improved, but loss of time increases due to additional processing overhead
Solution Approach 1:
The system implements feedback through acknowledgment mechanisms where receiving units send confirmation signals back to transmitting units. This feedback ensures reliable delivery by allowing retransmission of lost or corrupted packets, improving reliability while the time overhead is managed through efficient feedback protocols and parallel processing of multiple packets.
Data Source
AI summary
Embodiments of the present invention disclose a peripheral component interconnect express interface control unit. The unit includes a P2P module, configured to receive a first TLP from a RC or an EP and forward the first TLP to a reliable TLP transmission RTT module for processing; the reliable TLP transmission module, configured to determine, according to the received first TLP, sending links connected to active and standby PCIE switching units, and send the first TLP to the active and standby PCIE switching units through the sending links at the same time, so that a destination PCIE interface controller of the first TLP selectively receives the first TLP forwarded by the active and standby PCIE switching units and sends the first TLP to a destination EP or a destination RC, thereby implementing reliable transmission of a TLP in a case of a PCIE switching dual-plane networking connection.