Reliability improvement of distributed transaction processing optimizations based on connection status

Patent metadata

Field	Value
Publication number	US-9953053-B2
Application number	US-201414574918-A
Country	US
Kind code	B2
Filing date	Dec 18, 2014
Priority date	Dec 18, 2014
Publication date	Apr 24, 2018
Grant date	Apr 24, 2018

Abstract

A method, computer program product, and system for resolving a potential in-doubt condition of a distributed transaction, is provided. A processor receives a request to commit a transaction for a distributed transaction protocol that includes an applied process, the transaction includes a transfer of a commit decision from a coordinating node to a participating node. The processor checks the service status of the connection to the participating node, and finding the service status of the connection out of service or unavailable, the processor instructs the coordinating node to back-out (rollback) the transaction. Additionally, locality meta-data is used as an indication of reliability of the connection to the participating node, and in response to determining the participating node locality to be a remote network connection, the processor instructs the coordinating node to abort the applied process and send a standard distributed transaction protocol message over unreliable connections.

First claim

What is claimed is:

1. A computer program product for resolving a potential in-doubt condition of a distributed transaction, the computer program product comprising:
a computer readable storage medium having program instructions embodied therewith, wherein the program instructions are executable by a computer processor to cause the computer processor to perform a method comprising:
receiving a request to commit a transaction of a distributed transaction protocol that includes an applied process, the transaction includes at least a coordinating node and a participating node;
determining a service status of a connection to the participating node;
in response to determining the service status of the connection to the participating node to be unavailable, instructing the coordinating node to back-out the transaction;
determining a reliability of the connection of the participating node, based on meta-data corresponding to the participating node, wherein the meta-data indicates whether the locality of the participating node is a remote network connection, and whether the remote network connection includes a connection to a sub-network that is a different sub-network than that of the coordinating node of the transaction;
in response to determining the locality of the connection of the participating node is a remote network connection, and the locality of the remote network connection includes a connection to a sub-network that is a different sub-network than that of the coordinating node, communicating to the coordinating node to abort the applied process of the distributed transaction protocol and alternatively perform a standard two phase commit protocol process; and
in response to determining the locality of the connection of the participating node is a local network connection, transferring a decision to commit the transaction from the coordinating node to the participating node.

2. The computer program product of claim 1 , wherein the distributed transaction protocol is a two phase commit protocol and the applied process is an optimization process.

3. The computer program product of claim 2 wherein the optimization process is a last agent commit optimization process for a two phase commit protocol.

4. The computer program product of claim 1 , wherein determining the service status of the connection to the participating node is not available, is based on the service status of the connection that remains unavailable subsequent to sending a message, external to the transaction, to the participating node.

5. A computer system for resolving a potential in-doubt condition of a distributed transaction, the computer system comprising:
one or more computer processors;
one or more computer readable storage media; and
program instructions stored on the computer readable storage media for execution by at least one of the one or more processors, the program instructions comprising:
program instructions to receive a request to commit a transaction of a distributed transaction protocol that includes an applied process, the transaction includes at least a coordinating node and a participating node;
program instructions to determine, by the processor, a service status of a connection to the participating node; and
in response to determining the service status of the connection to the participating node to be unavailable, program instructions to instruct the coordinating node to back-out the transaction;
program instructions to determine a reliability of the connection of the participating node, based on meta-data corresponding to the participating node, wherein the meta-data indicated whether the locality of the participating node is a remote network connection, and whether the remote network connection includes a connection to a sub-network that is a different sub-network than that of the coordinating node of the transaction;
in response to determining the locality of the connection of the participating node is a remote network connection, and the locality of the remote network connection includes a connection to a sub-network that is a different sub-network than that of the coordinating node, program instructions to communicate to the coordinating node to abort the applied process of the distributed transaction protocol and alternatively perform a standard two phase commit protocol process; and
in response to determining the locality of the connection of the participating node is a local network connection, program instructions to transfer a decision to commit the transaction from the coordinating node to the participating node.

6. The computer system of claim 5 , wherein the distributed transaction protocol is a two phased commit protocol, and the applied process is a last agent commit optimization process for the two phase commit protocol and the at least one participating node is a last agent of the last agent commit optimization process.

7. The computer system of claim 5 , wherein determining the service status of the connection to the participating node is not available, is based on the service status of the connection that remains unavailable subsequent to sending a message, external to the transaction, to the participating node.

Assignees

• IBM

Inventors

• Brooks Michael D
• Mitchell Ian J
• Wakelin Philip I

Classifications

• H04L41/0836

  to enhance reliability, e.g. reduce downtime · CPC title

• G06F17/30377 Primary

  Physics · mapped topic

• H04L67/1012

  based on compliance of requirements or conditions with available server resources · CPC title

• H04L51/24

  Electricity · mapped topic

• H04L67/1025

  Dynamic adaptation of the criteria on which the server selection is based · CPC title