Communication between multi-processor clusters of multi-cluster computer systems
View Patent ↗Improved techniques are provided for detecting and correcting errors and skew in inter-cluster communications within computer systems having a plurality of multi-processor clusters. The local nodes of each cluster include a plurality of processors and an interconnection controller. Intra-cluster links are formed between the local nodes, including the interconnection controller, within a cluster. Inter-cluster links are formed between interconnection controllers of different clusters. Intra-cluster packets may be serialized and encapsulated as inter-cluster packets for transmission on inter-cluster links, preferably with link-layer encapsulation. Each inter-cluster packet may include a sequence identifier and error information computed for that packet. Clock data may be embedded in symbols sent on each bit lane of the inter-cluster links. Copies of transmitted inter-cluster packets may be stored until an acknowledgement is received. The use of inter-cluster packets on an inter-cluster link is preferably transparent to other links and to the protocol layer.
1. A computer system comprising a plurality of processor clusters interconnected by a plurality of point-to-point inter-cluster links, each processor cluster comprising nodes including:
a plurality of local processors; and
an interconnection controller;
wherein the nodes are interconnected by a plurality of point-to-point intra-cluster links,
wherein communications within a cluster are made via an intra-cluster protocol that uses intra-cluster packets,
wherein the interconnection controller in each cluster is operable to map locally-generated communications directed to others of the clusters to the point-to-point inter-cluster links and to map remotely-generated communications directed to the local nodes to the point-to-point intra-cluster links,
wherein communications between clusters are made via an inter-cluster protocol that uses inter-cluster packets, an inter-cluster packet encapsulating at least one intra-cluster packet, and
wherein each interconnection controller is configured to compute a first cyclic redundancy code check for, and to encode first cyclic redundancy code check data in, each inter-cluster packet transmitted on the point-to-point inter-cluster links.
2. The computer system of claim 1 , wherein each interconnection controller is further configured to compute a second cyclic redundancy code check for each inter-cluster packet received on the point-to-point inter-cluster links.
3. The computer system of claim 1 , wherein each interconnection controller is further configured to transmit each inter-cluster packet as an indivisible unit on the point-to-point inter-cluster links.
4. The computer system of claim 1 , wherein each interconnection controller is further configured to encode a sequence identifier in each inter-cluster packet transmitted on the point-to-point inter-cluster links.
5. The computer system of claim 1 , wherein the nodes are configured to compute a cyclic redundancy code check on a transmission window that includes transmissions of multiple intra-cluster packets on the point-to-point intra-cluster links.
6. The computer system of claim 1 , wherein each interconnection controller is further configured to compute a cyclic redundancy code check on a transmission window that includes transmissions of multiple intra-cluster packets on the point-to-point intra-cluster links.
7. The computer system of claim 1 , wherein each interconnection controller encodes the cyclic redundancy code check data in a field reserved for a link layer of each inter-cluster packet transmitted on the point-to-point inter-cluster links.
8. A computer system, comprising:
a first cluster including a first plurality of processors and a first interconnection controller, the first plurality of processors and the first interconnection controller interconnected by first point-to-point intra-cluster links; and
a second cluster including a second plurality of processors and a second interconnection controller, the second plurality of processors and the second interconnection controller interconnected by second point-to-point intra-cluster links, the first interconnection controller coupled to the second interconnection controller by point-to-point inter-cluster links;
wherein the first interconnection controller is configured to:
receive a request according to an intra-cluster protocol from a first processor in the first plurality of processors;
generate a probe packet responsive to the request;
compute a first cyclic redundancy code check based only upon bits in the probe packet;
encode first cyclic redundancy code check data in the probe packet; and
send the probe packet to the second interconnection controller in the second cluster.
9. The computer system of claim 8 , wherein the first interconnection controller is further configured to encode a sequence identifier in the probe packet.
10. The computer system of claim 8 , wherein the second interconnection controller is configured to compute a second cyclic redundancy code check based only upon bits in the first probe packet.
11. The computer system of claim 8 , wherein the second interconnection controller is configured to forward the probe packet according to the intra-cluster protocol to a processor in the second plurality of processors.
12. The computer system of claim 8 , wherein the second interconnection controller is configured to forward the probe packet according to the intra-cluster protocol to each processor in the second plurality of processors.
13. The computer system of claim 11 , wherein the processor in the second plurality of processors is configured to send a response packet according to the intra-cluster protocol to the second interconnection controller.
14. The computer system of claim 12 , wherein each processor in the second plurality of processors is configured to send a response packet according to the intra-cluster protocol to the second interconnection controller.
15. An interconnection controller, comprising:
an intra-cluster interface configured for coupling with intra-cluster links to a plurality of local processors arranged in a point-to-point architecture in a local cluster;
an inter-cluster interface configured for coupling with an inter-cluster link to a non-local interconnection controller in a non-local cluster;
a transceiver configured to receive an intra-cluster packet from a local processor via an intra-cluster link and encode a sequence identifier in a header of the intra-cluster packet; and
a serializer/deserializer configured to serialize the encoded packet and forward the encoded, serialized packet to the inter-cluster interface for transmission to the non-local interconnection controller via an inter-cluster link.
16. The interconnection controller of claim 15 , wherein the transceiver is further configured to compute a cyclic redundancy code check based only on the encoded packet and to encode the cyclic redundancy code check in the encoded packet.
17. The interconnection controller of claim 15 , wherein the inter-cluster interface is further configured to receive encoded, serialized packets from the non-local interconnection controller, wherein the serializer/deserializer is further configured to deserialize the encoded, serialized packets and wherein the transceiver is further configured to perform a cyclic redundancy code check on the deserialized packets.
18. An integrated circuit comprising the interconnection controller of claim 15 .
19. A set of semiconductor processing masks representative of at least a portion of the interconnection controller of claim 15 .
20. At least one computer-readable medium having data structures stored therein representative of the interconnection controller of claim 15 .
21. The integrated circuit of claim 18 , wherein the integrated circuit comprises an application-specific integrated circuit.
22. The at least one computer-readable medium of claim 20 , wherein the data structures comprise a simulatable representation of the interconnection controller.
23. The at least one computer-readable medium of claim 20 , wherein the data structures comprise a code description of the interconnection controller.
24. The at least one computer-readable medium of claim 22 , wherein the simulatable representation comprises a netlist.
25. The at least one computer-readable medium of claim 23 , wherein the code description corresponds to a hardware description language.
26. A computer-implemented method for detecting errors in a computer system comprising a plurality of clusters, each cluster including a plurality of local nodes and an interconnection controller interconnected by point-to-point intra-cluster links, communications between the local nodes and the interconnection controller made via an intra-cluster protocol using intra-cluster packets, the interconnection controller of each cluster interconnected by point-to-point inter-cluster links with the interconnection controller of other clusters, the computer-implemented method comprising:
forming an inter-cluster packet by encapsulating an intra-cluster packet;
encoding a sequence identifier in the inter-cluster packet;
calculating first cyclic redundancy code check data based only upon the inter-cluster packet;
encoding the first cyclic redundancy code check data in the inter-cluster packet; and
transmitting the inter-cluster packet from a first interconnection controller to a second interconnection controller on a point-to-point inter-cluster link.
27. The computer-implemented method of claim 26 , wherein the encoding steps comprise encoding in an area of the inter-cluster packet reserved for link layer information.
28. The computer-implemented method of claim 26 , further comprising:
receiving the inter-cluster packet; and
calculating second cyclic redundancy code check data based only upon the inter-cluster packet.
29. The computer-implemented method of claim 28 , further comprising:
detecting an error in the inter-cluster packet based upon the second cyclic redundancy code check data; and
notifying the first interconnection controller of the error.
30. An apparatus for detecting errors in a computer system comprising a plurality of clusters, each cluster including a plurality of local nodes and an interconnection controller interconnected by point-to-point intra-cluster links, communications between the local nodes and the interconnection controller made via an intra-cluster protocol using intra-cluster packets, the interconnection controller of each cluster interconnected by point-to-point inter-cluster links with the interconnection controller of other clusters, the apparatus comprising:
means for forming an inter-cluster packet by encapsulating an intra-cluster packet;
means for encoding a sequence identifier in the inter-cluster packet;
means for calculating first cyclic redundancy code check data based only upon the inter-cluster packet;
means for encoding the first cyclic redundancy code check data in the inter-cluster packet; and
means for transmitting the inter-cluster packet from a first interconnection controller to a second interconnection controller on a point-to-point inter-cluster link, wherein the encoding means comprise means for encoding in an area of the inter-cluster packet reserved for link layer information.