Method for switching a node controller link, processor system, and node

1. A method for switching a node controller link, wherein a processor system comprises more than two nodes capable of communicating with each other, each node comprises one node controller NC chip, one host bus adapter HBA apparatus, and at least one CPU, the NC chip is connected to each CPU in a node where the NC chip is located, the HBA apparatus is connected to each CPU in a node where the HBA apparatus is located, and an NC link borne by the NC chip corresponds to an HBA link borne by the HBA apparatus; the method comprises:

when it has been detected that a failure occurs in an NC chip, switching a service on an NC link borne by the NC chip to a corresponding HBA link;

wherein that an NC link borne by the NC chip that corresponds to an HBA link borne by the HBA apparatus comprises: presetting a first routing table and a second routing table in each node, wherein the first routing table is a routing table of an NC chip in each node and each NC chip corresponds to an address of a node where the NC chip is located, the second routing table is a routing table of an HBA apparatus in each node and each HBA apparatus corresponds to an address of a node where the HBA apparatus is located, and the first routing table corresponds to the second routing table through a node address.

2. The method according to claim 1 , wherein the switching a service on an NC link to a corresponding HBA link comprises:

searching, by using the first routing table, for an address of a node where a NC chip in which the failure occurs is located;

searching, by using the second routing table, for an HBA apparatus corresponding to the address; and

switching a service on the NC link borne by the NC chip in which the failure occurs to the corresponding HBA link borne by the HBA apparatus.

3. The method according to claim 1 , further comprising: when an operating system detects that a bandwidth occupation rate of an NC link on a node exceeds a threshold, notifying the node to switch a service conforming to a preset table from the NC link to a corresponding HBA link.

4. The method according to claim 1 , wherein a manner in which the HBA apparatus is connected to each CPU in a node where the HBA apparatus is located is: the HBA apparatus is inserted in a north bridge chip, and the north bridge chip is connected to each CPU through a front side bus.

5. The method according to claim 1 , wherein a manner in which the HBA apparatus is connected to each CPU in a node where the HBA apparatus is located is: the HBA apparatus is directly mounted on each CPU.

6. A processor system, comprising more than two nodes capable of communicating with each other, wherein

each node comprises a node controller NC chip, a host bus adapter HBA apparatus, and at least one CPU, the NC chip is connected to each CPU in a node where the NC chip is located, the HBA apparatus is connected to each CPU in a node where the HBA apparatus is located, and an NC link borne by the NC chip corresponds to an HBA link borne by the HBA apparatus; and

when it has been detected that a failure occurs in the NC chip of the node itself, the node switches a service on an NC link borne by the NC chip to a corresponding HBA link;

wherein the node is configured to:

store a first routing table and a second routing table which are preset, wherein the first routing table is a routing table of an NC chip in each node and each NC chip corresponds to an address of a node where the NC chip is located, the second routing table is a routing table of an HBA apparatus in each node and each HBA apparatus corresponds to an address of a node where the HBA apparatus is located, and the first routing table corresponds to the second routing table through a node address.

7. The processor system according to claim 6 , wherein the node is further configured to:

switch a service that is on an NC link and conforms to a preset table to a corresponding HBA link, when receiving a notification of an operating system about switching the NC link of this node.

8. A node, comprising:

a node controller NC chip, a host bus adapter HBA apparatus, a controller, and at least one CPU, wherein the NC chip is connected to each CPU in a node where the NC chip is located, the HBA apparatus is connected to each CPU in a node where the HBA apparatus is located, and an NC link borne by the NC chip corresponds to an HBA link borne by the HBA apparatus; and

the controller is configured to, when it has been detected that a failure occurs in the NC chip of the node itself, switch a service on an NC link borne by the NC chip to a corresponding HBA link;

wherein the node further comprising a storage device, configured to store a first routing table and a second routing table which are preset, wherein the first routing table is a routing table of an NC chip in each node and each NC chip corresponds to an address of a node where the NC chip is located, the second routing table is a routing table of an HBA apparatus in each node and each HBA apparatus corresponds to an address of anode where the HBA apparatus is located, and the first routing table corresponds to the second routing table through a node address.

9. The node according to claim 8 , wherein the controller is further configured to:

switch a service that is on the NC link and conforms to a preset table to a corresponding HBA link, when receiving a notification of an operating system about switching the NC link of this node.