Managing membership of a domain of processors in a distributed computing environment

• US 5,896,503 A
• Filed: 07/23/1996
• Issued: 04/20/1999
• Est. Priority Date: 07/23/1996
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Brenner, Larry Bert, Kirby, Orvalle Theodore, Chandra, Tushar Deepak, Pershing, John Arthur Jr., Badovinatz, Peter Richard
• Primary Examiner(s): Rinehart, Mark H.
• Assistant Examiner(s): PIERCE, IVAN
• Application Number:
  US08/686,233
• Time in Patent Office: 1,001 Days
• Field of Search: 395/200.3, 395/200.31, 395/200.35, 395/182.02
• US Class Current: 709/201
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06F 11/1425 by reconfiguration of node membership

Abstract

Managing membership of a domain of processors in a distributed computing environment including determining the lowest numbered node available in the domain and periodically sending grovel messages from the other available nodes to the lowest numbered node in the domain. Responsive to receiving all of the grovel messages by the lowest numbered node, the lowest numbered node performs a two phase commit whereby the lowest numbered node becomes a name server for managing the processors in the domain. The two phase commit includes sending an insert phase 1 (P1) message to the other available nodes from the lowest numbered node and, responsive to the insert P1 message, sending an acknowledge (ACK) message from the other available nodes to the lowest numbered node. Responsive to the ACK messages, a commit message is sent to the other available nodes notifying them that the lowest numbered node has become the name server.

Claims

1. A method for managing membership of a domain of processors in a distributed computing environment, said method comprising the steps of:

determining the lowest numbered node available in the domain;

periodically sending grovel messages from the other available nodes to the lowest numbered node in the domain;

responsive to receiving all of said grovel messages by said lowest numbered node, performing a two phase commit protocol whereby the lowest numbered node becomes a name server for managing said processors in said domain.

2. The method of claim 1 whereby the two phase commit protocol comprises the steps of:

sending an insert phase 1 (P1) message to said other available nodes from the lowest numbered node;

responsive to said insert P1 message, sending an acknowledge (ACK) message from said other available nodes to said lowest numbered node; and

responsive to said ACK messages, sending a commit message to said other available nodes notifying them that said lowest numbered node has become the name server.

3. The method of claim 1 comprising the steps of:

timing the receipt of said grovel messages for a set period of time; and

receiving said grovel messages only during said set period of time.

4. The method of claim 2 comprising the steps of:

responsive to receipt by the lowest numbered node of a grovel message wherein said lowest numbered node has a reputed name server (RNS), sending a my leader is (MLI) message to nodes sending grovel messages, said MLI messages identifying said RNS;

responsive to the receipt of said MLI message, periodically sending grovel messages to said RNS from said nodes receiving said MLI message; and

responsive to said grovel messages, said RNS performing said two phase commit to establish said RNS as the name server.

5. An apparatus for managing membership of a domain of processors in a distributed computing environment, said apparatus comprising:

means for determining the lowest numbered node available in the domain;

grovel message means for periodically sending grovel messages from the other available nodes to the lowest numbered node in the domain;

two phase commit means responsive to receiving all of said grovel messages by said lowest numbered node, said two phase commit means performing a two phase commit whereby the lowest numbered node becomes a name server for managing said processors in said domain.

6. The apparatus of claim 5 whereby the two phase commit means comprises:

means for sending an insert phase 1 (P1) message to said other available nodes from the lowest numbered node;

acknowledge means responsive to said insert P1 message, said acknowledge means sending an acknowledge (ACK) message from said other available nodes to said lowest numbered node; and

commit message means responsive to said ACK messages, said commit messag means sending a commit message to said other available nodes notifying them that said lowest numbered node has become the name server.

7. The apparatus of claim 5 comprising:

a timer for timing the receipt of said grovel messages for a set period of time, said two phase commit means receiving said grovel messages only during said set period of time.

8. The apparatus of claim 6 comprising:

MLI message means responsive to receipt by the lowest numbered node of a grovel message wherein said lowest numbered node has a reputed name server (RNS), said MLI message means sending a my leader is (MLI) message to nodes sending grovel messages, said MLI messages identifying said RNS;

second grovel message means responsive to the receipt of said MLI message, said second grovel message means periodically sending grovel messages to said RNS from said nodes receiving said MLI message; and

second two phase commit means in said RNS responsive to said grovel messages, said second two phase commit means in said RNS performing said two phase commit to establish said RNS as the name server.

9. A program product recorded on a computer readable medium, said program product for managing membership of a domain of processors in a distributed computing environment, said program product comprising:

determining the lowest numbered node available in the domain;

periodically sending grovel messages from the other available nodes to the lowest numbered node in the domain;

responsive to receiving all of said grovel messages by said lowest numbered node, performing a two phase commit protocol whereby the lowest numbered node becomes a name server for managing said processors in said domain.

10. The program product of claim 9 whereby the two phase commit protocol comprises the steps of:

sending an insert phase 1 (P1) message to said other available nodes from the lowest numbered node;

responsive to said insert P1 message, sending an acknowledge (ACK) message from said other available nodes to said lowest numbered node; and

responsive to said ACK messages, sending a commit message to said other available nodes notifying them that said lowest numbered node has become the name server.

11. The program product of claim 9 comprising:

timing the receipt of said grovel messages for a set period of time; and

receiving said grovel messages only during said set period of time.

12. The program product of claim 10 comprising:

responsive to receipt by the lowest numbered node of a grovel message wherein said lowest numbered node has a reputed name server (RNS), sending a my leader is (MLI) message to nodes sending grovel messages, said MLI messages identifying said RNS;

responsive to the receipt of said MLI message, periodically sending grovel messages to said RNS from said nodes receiving said MLI message; and

responsive to said grovel messages, said RNS performing said two phase commit to establish said RNS as the name server.