Method and apparatus for synchronizing nodes in massively parallel systems

• US 6,418,526 B1
• Filed: 11/15/1999
• Issued: 07/09/2002
• Est. Priority Date: 11/15/1999
• Status: Expired due to Term

• Current Assignee: Teradata US, Inc. (Teradata Corporation)
• Original Assignee: NCR Corporation (NCR Voyix Corporation)
• Inventors: Merritt, John E., Denman, Robert W.
• Primary Examiner(s): Follansbee, John A.
• Application Number:
  US09/440,353
• Time in Patent Office: 967 Days
• Field of Search: 712/10-22, 713/400
• US Class Current: 712/16
• CPC Class Codes: G06F 2209/505 Clust
• CPC Class Codes: G06F 9/5061 Partitioning or combining of resources

Abstract

A method, apparatus, article of manufacture, and a memory stricture for synchronizing nodes in configuring a massively parallel processing system is disclosed. The method comprises the steps of receiving a registration request from each of the non-coordinator nodes in the coordinator node and, after a registration request has been received from all of the non-coordinator nodes, sending a wakeup message from the coordinator node to each of the non-coordinator nodes to wake the non-coordinator nodes from a sleep state. In another embodiment of the invention, the method comprises the steps of transmitting a registration request from a non-coordinator node to a coordinator node; commanding the non-coordinator node to enter a sleep state; receiving a multicast wakeup message from the coordinator node in the non-coordinator node after a registration request is received from all of the non-coordinator nodes in the coordinator node; and waking the non-coordinator node in response to the wakeup message. The apparatus comprises a means for performing the steps described above, and the article of manufacture comprises a program storage device tangibly embodying computer instructions for performing the above method steps.

Claims

1. A method of synchronizing a parallel processing system comprising a plurality of nodes including a coordinator node and at least one non-coordinator node, comprising the steps of:

receiving a registration request from each of the non-coordinator nodes in the coordinator node; and

sending a wakeup request from the coordinator node to each of the non-coordinator nodes to wake the non-coordinator nodes from a sleep state after a registration request has been received from all of the non-coordinator nodes.

2. The method of claim 1, wherein each non-coordinator node is associated with a node ID, and wherein;

the step of receiving a registration request from each of the non-coordinator nodes in the coordinator node comprises the step of adding the node ID from each non-coordinator node to a node synchronization list and incrementing a node synchronization count; and

the step of sending a wakeup message from the coordinator node to each of the non-coordinator nodes comprises the step of broadcasting a wakeup message from the coordinator node to each of the non-coordinator nodes from which a registration request was received after the node synchronization counts indicates that a registration request was received from all non-coordinator nodes.

3. A method of synchronizing a parallel processing system comprising a plurality of nodes including a coordinator node and at least one non-coordinator node, comprising the steps of:

transmitting a registration request from a non-coordinator node to the coordinator node;

commanding the non-coordinator node to a sleep state;

receiving a multicast wakeup message from the coordinator node in the non-coordinator node after a registration request is received from all of the non-coordinator nodes in the coordinator node; and

waking the non-coordinator node in response to the wakeup message.

4. The method of claim 3, further comprising the steps of:

setting a wakeup timer after commanding the non-coordinator node to a sleep state; and

waking the non-coordinator node when the wakeup timer has expired.

5. An apparatus for synchronizing a parallel processing system comprising a plurality of nodes including a coordinator node and at least one non-coordinator node, the apparatus comprising:

means for receiving a registration request from each of the non-coordinator nodes in the coordinator node; and

means for sending a wakeup request from the coordinator node to each of the non-coordinator nodes to wake the non-coordinator nodes from a sleep state after a registration request has been received from all of the non-coordinator nodes.

6. The apparatus of claim 5, wherein each non-coordinator node is associated with a node ID, and wherein:

the means for receiving a registration request from each of the non-coordinator nodes in the coordinator node comprises means for adding the node ID from each non-coordinator node to a node synchronization list and incrementing a node synchronization count; and

the means for sending a wakeup message from the coordinator node to each of the non-coordinator nodes comprises means for broadcasting a wakeup message from the coordinator node to each of the non-coordinator nodes from which a registration request was received after the node synchronization counts indicates that a registration request was received from all non-coordinator nodes.

7. An apparatus for synchronizing a parallel processing system comprising a plurality of nodes including a coordinator node and at least one non-coordinator node, the apparatus comprising:

means for transmitting a registration request from a non-coordinator node to the coordinator node;

means for commanding the non-coordinator node to a sleep state;

means for receiving a multicast wakeup message from the coordinator node in the non-coordinator node after a registration request is received from all of the non-coordinator nodes in the coordinator node; and

means for waking the non-coordinator node in response to the wakeup message.

8. The apparatus of claim 7, further comprising:

means for setting a wakeup timer after commanding the non-coordinator node to a sleep state; and

means for waking the non-coordinator node when the wakeup timer has expired.

9. A program storage device, readable by a computer, tangibly embodying at least one program of instructions executable by the computer to perform method steps of synchronizing a parallel processing system comprising a plurality of nodes including a coordinator node and at least one non-coordinator node, the method steps comprising the steps of:

receiving a registration request from each of the non-coordinator nodes in the coordinator node; and

sending a wakeup request from the coordinator node to each of the non-coordinator nodes to wake the non-coordinator nodes from a sleep state after a registration request has been received from all of the non-coordinator nodes.

10. The program storage device of claim 9, wherein each non-coordinator node is associated with a node ID, and wherein:

the method step of receiving a registration request from each of the non-coordinator nodes in the coordinator node comprises the method step of adding the node ID from each non-coordinator node to a node synchronization list and incrementing a node synchronization count; and

the method step of sending a wakeup message from the coordinator node to each of the non-coordinator nodes comprises the method step of broadcasting a wakeup message from the coordinator node to each of the non-coordinator nodes from which a registration request was received after the node synchronization counts indicates that a registration request was received from all non-coordinator nodes.

11. A program storage device, readable by a computer, tangibly embodying at least one program of instructions executable by the computer to perform method steps of synchronizing a parallel processing system comprising a plurality of nodes including a coordinator node and at least one non-coordinator node, the method steps comprising the steps of:

transmitting a registration request from a non-coordinator node to the coordinator node;

commanding the non-coordinator node to a sleep state;

receiving a multicast wakeup message from the coordinator node in the non-coordinator node after a registration request is received from all of the non-coordinator nodes in the coordinator node; and

waking the non-coordinator node in response to the wakeup message.

12. The program storage device of claim 11, wherein the method steps further comprise the steps of:

setting a wakeup timer after commanding the non-coordinator node to a sleep state; and

waking the non-coordinator node when the wakeup timer has expired.