HOP-BY-HOP ERROR DETECTION IN A SERVER SYSTEM

• US 20140189443A1
• Filed: 12/31/2012
• Published: 07/03/2014
• Est. Priority Date: 12/31/2012
• Status: Active Grant

• Current Assignee: Advanced Micro Devices, Inc.
• Original Assignee: Advanced Micro Devices, Inc.
• Inventors: Xu, Min, Lie, Sean, Shen, Gene
• Granted Patent: US 9,176,799 B2
• Time in Patent Office: Days
• US Class Current: 714/48
• CPC Class Codes: G06F 11/0724 in a multiprocessor or a multi-core unit multiprocessors per se G06F15/80
• CPC Class Codes: G06F 11/079 Root cause analysis, i.e. error or fault diagnosis in a hardware test environment G06F11/22; in a software test environment G06F11/36
• CPC Class Codes: G06F 11/08 Error detection or correction by redundancy in data representation, e.g. by using checking codes
• CPC Class Codes: G06F 11/1004 to protect a block of data words, e.g. CRC or checksum G06F11/1076 takes precedence; security arrangements for protecting computers or computer systems against unauthorized activity G06F21/00
• CPC Class Codes: H04L 1/00 Arrangements for detecting or preventing errors in the information received correcting synchronisation H04L7/00
• CPC Class Codes: H04L 1/0061 Error detection codes
• CPC Class Codes: H04L 2001/0097 Relays
• CPC Class Codes: H04L 43/065 related to network devices
• CPC Class Codes: H04L 43/0823 Errors, e.g. transmission errors
• CPC Class Codes: H04L 47/10 Flow control; Congestion control

Abstract

A server system performs error detection on a hop-by-hop basis at multiple compute nodes, thereby facilitating the detection of a compute node experiencing failure. The server system communicates a packet from an originating node (the originating node) to a destination node by separating the packet into multiple flow control digits (flits) and routing the flits using a series of hops over a set of intermediate nodes. The packet'"'"'s final flit includes error detection information, such as checksum data. As each intermediate node receives the final flit, it performs error detection using the error detection information. The pattern of nodes that detect an error indicates which intermediate node has experienced a failure.

Claims

1. A server system comprising:

a network of nodes coupled in a network topology via a routing fabric, wherein;

the nodes are to communicate a packet from an originating node to a destination node via a set of intermediate nodes; and

each intermediate node of the set of intermediate nodes is to perform error detection on the packet to produce a plurality of error reports.

2. The server system of claim 1, wherein the originating node is to separate the packet into a set of flow control digits (flits), and where each intermediate node of the set of intermediate nodes is to initiate error detection in response to receiving a first flit of the set of flits and complete error detection in response to receiving a last flit of the set of flits, the error detection based on error detection information generated based on all of the flits for the packet.

3. The server system of claim 1, wherein the server system is to identify a faulty node of the set of intermediate nodes based on the plurality of error reports.

4. The server system of claim 3, wherein the set of intermediate nodes form a routing path between the originating node and the destination node and the server system is to identify the faulty node by identifying the faulty node as a last intermediate node in the routing path whose corresponding error report does not indicate an error.

5. The server system of claim 3, wherein the server system is to identify the faulty node in response to detecting an error in the packet at the destination node.

6. The server system of claim 3, wherein the server system is to change a routing scheme of the routing fabric in response to identifying the faulty node, such that faulty node is not used in the routing scheme as an intermediate node to route packets between originating nodes and destination nodes.

7. The server system of claim 1, wherein the network topology comprises a three-dimensional network topology.

8. The server system of claim 1, wherein the originating node comprises a compute node that communicates with the routing fabric that virtualizes the routing fabric so that it appears to a processor of the compute node as a network interface device.

9. The server system of claim 1, wherein the originating node comprises a compute node that communicates with the routing fabric that virtualizes the routing fabric so that it appears to a processor of the compute node as a storage device.

10. A server system, comprising:

a plurality of field replaceable units (FRUs) each of the plurality of field replaceable units comprising a corresponding node to execute services for the server system; and

a routing fabric configured to implement a network topology to route a packet between an originating node and a destination node via a set of intermediate nodes, wherein each of the intermediate nodes of the set of intermediate nodes is to perform error detection on the packet as it is communicated.

11. The server system of claim 10, wherein the originating node is to separate the packet into a set of flow control digits (flits), and where each intermediate node of the set of intermediate nodes is to initiate error detection in response to receiving a first flit of the set of flits and complete error detection in response to receiving a last flit of the set of flits.

12. The server system of claim 10, wherein the server system is to identify a faulty node of the set of intermediate nodes based on the error detection at each of the intermediate nodes.

13. The server system of claim 12, wherein the server system is to identify an FRU corresponding to the faulty node based on the error detection at each of the intermediate nodes.

14. The server system of claim 10, further comprising a first FRU including a network interface to provide the server system with an interface to an external network.

15. The server system of claim 14, further comprising a second FRU including a storage interface to provide the server system with an interface to a storage device.

16. The server system of claim 10, wherein the network topology comprises a three-dimensional torus topology.

17. A method, comprising:

communicating a packet from a first node of a server system to a second node of the server system via a set of intermediate nodes coupled to a routing fabric of the server system; and

performing error detection on the packet at each of the intermediate nodes.

18. The method of claim 17, further comprising:

separating the packet into a set of flow control digits (flits) at the first node; and

wherein performing error detection on the packet comprises;

at each of the intermediate nodes, initiating error detection in response to receiving a first flit of the set of flits and completing error detection in response to receiving a last flit of the set of flits.

19. The method of claim 17, further comprising:

generating an error report at each of the intermediate nodes based on the error detection to generate a plurality of error reports; and

identifying at the server system a faulty node of the set of intermediate nodes based on the plurality of error reports.

20. The method of claim 19, wherein the set of intermediate nodes form a routing path between the first node and the second node and identifying the faulty node comprises identifying the faulty node as a last intermediate node in the routing path whose corresponding error report does not indicate an error.

21. The method of claim 17, further comprising changing a routing scheme for the server system in response to detecting an error based on the error detection.