Chiplet Protocol Interface Path Field Ordering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Chiplet systems face limitations in supporting multiple communication standards and protocols, leading to restricted applicability when different device vendors use different packet communication standards, necessitating a solution for ordered delivery of data packets across chiplets with varying path information types.
Innovation Solution
The implementation of a chiplet protocol interface (CPI) network that uses a path field and a bridge-type field in data packets to indicate the type of path information, enabling both address-based and transaction identifier-based ordering, allowing for flexible communication across chiplets using different standards and protocols.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If chiplet systems use a single communication protocol, then device complexity is reduced, but adaptability to different vendors' standards deteriorates
Solution Approach 1:
The CPI network interface is designed to support multiple communication protocols and standards simultaneously. The path field can accommodate different types of path information (destination address, transaction identifier, or other routing data) depending on the protocol being used, allowing the same physical interface to communicate with chiplets from different vendors that may use different communication standards.
Solution Approach 2:
The system changes the interpretation of the path field parameters based on the protocol type. The same physical field can represent different logical concepts (destination address in one protocol, transaction identifier in another), allowing flexible adaptation to different vendors' standards without changing the physical interface architecture.
2Measurement precision
If data packets include detailed path information for precise routing, then routing precision is improved, but device complexity increases due to multiple protocol support requirements
Solution Approach 1:
The path field serves multiple functions depending on the protocol context. It can carry destination addresses for address-based routing, transaction identifiers for ordering, or other routing information, all within the same physical field structure. This universal design provides routing precision for different protocols without requiring separate fields for each protocol type.
Solution Approach 2:
The data packet structure is segmented into distinct fields (path field, bridge-type field, etc.) that can be independently interpreted based on the protocol type. This segmentation allows the path field to be precisely interpreted according to the specific protocol being used, maintaining routing precision while simplifying protocol handling through structured data organization.
3Adaptability or versatility
If the system supports multiple path information types in packets, then adaptability to different communication standards is improved, but difficulty of detecting and measuring path information type increases
Solution Approach 1:
The bridge-type field acts as an intermediary that explicitly indicates the type of path information contained in the path field. This mediator field resolves the ambiguity of detecting path information type by providing a clear, standardized indicator that tells the receiving device how to interpret the path field, whether it contains a destination address, transaction identifier, or other routing data.
Solution Approach 2:
The system uses parameter changes in the bridge-type field to indicate different path information types. By encoding the path information type as a distinct parameter (bridge-type), the system makes it easy to detect and measure the type of path information without complex analysis of the path field content itself.
4Speed
If data packets are ordered based on destination address, then routing efficiency is improved, but loss of time occurs for packets requiring transaction identifier-based ordering
Solution Approach 1:
The ordering mechanism is dynamic and adaptable based on the bridge-type field indication. The system can switch between address-based ordering and transaction identifier-based ordering depending on the protocol type and requirements. This dynamic approach allows routing optimization for address-based packets while providing appropriate transaction-based ordering when needed, minimizing overall time loss.
Solution Approach 2:
The ordering parameter changes based on the bridge-type field. When the bridge-type indicates address-based routing, the system uses destination address for ordering to achieve routing efficiency. When the bridge-type indicates transaction identifier-based routing, the system switches to using transaction identifiers for ordering. This parameter change allows the system to optimize for speed when appropriate while maintaining correctness for transaction-ordered packets.
Data Source
AI summary
A system may include multiple electronic devices and multiple hardware transceivers. The multiple electronic devices may be coupled to each other via an interface network, and may include multiple chiplets. The multiple hardware transceivers, with at least one transceiver included in or coupled to a respective electronic device of the multiple electronic devices, may each be configured to receive data packets from a source device. The data packets may each include a path field including path information indicating a path to a destination device and a bridge-type field including bridge-type information indicating a type of the path information in the path field. The source device and the destination device may each include a chiplet. The multiple hardware transceivers may each be further configured to transmit the received data packets to the destination device using the path information and the bridge-type information of each received data packet.