Migration method for software application in a multi-computing architecture, method for carrying out functional continuity implementing said migration method and multi-computing system provided therewith

• US 7,577,720 B2
• Filed: 07/28/2003
• Issued: 08/18/2009
• Est. Priority Date: 08/02/2002
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: International Business Machines Corporation
• Inventors: Dufour, Laurent, Vertes, Marc
• Primary Examiner(s): Cardone; Jason
• Assistant Examiner(s): Mirza; Adnan M
• Application Number:
  US10/523,025
• Publication Number:
  US 20050262411A1
• Time in Patent Office: 2,213 Days
• Field of Search: 709/209, 709/219, 709/223, 370/229
• US Class Current: 709/219
• CPC Class Codes: G06F 11/1482 by means of middleware or OS functionality
• CPC Class Codes: G06F 11/1658 Data re-synchronization of a redundant component, or initial sync of replacement, additional or spare unit
• CPC Class Codes: G06F 11/203 using migration
• CPC Class Codes: G06F 11/2033 switching over of hardware resources
• CPC Class Codes: G06F 11/2041 with more than one idle spare processing component
• CPC Class Codes: G06F 11/2097 maintaining the standby controller/processing unit updated initialisation or re-synchronisation thereof G06F11/1658 and subgroups
• CPC Class Codes: G06F 2201/815 Virtual middleware or OS functionality using virtual machines to implement generic software techniques for error detection or fault masking G06F11/1484
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1095 Replication or mirroring of data, e.g. scheduling or transport for data synchronisation between network nodes
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]

Abstract

A method for carrying out a connection migration in a multicomputing architecture (cluster), from a first node which is called a primary node or operation node, including a first computer of the cluster on which an initial software application is run, towards at lest one secondary node including another computer of the cluster. The method implements a virtual network address which is carried by the first computer and transferred to the other computer, the virtual network address being embodied in the form of a dialogue link between the cluster and client computers connected thereto and associated with the software application. The associated connections can, for example be associated with a software application which is replicated in another computer in such a way that the initial application service is switched to the replica thereof.

Claims

1. A method for carrying out a connection migration and functional continuity of a software application in a multi-computer architecture, the method comprising:

executing, from a first node, an initial software application on a first computer in the multi-computer architecture, wherein the first node is a primary node, and wherein the initial software application is executed towards at least one secondary node, wherein the at least one secondary node is on a second computer in the multi-computer architecture;

implementing a virtual network address which is carried by the first computer;

transferring, by the first computer, the virtual network address to the second computer;

providing the virtual network address as a dialogue link to client computers connected to the multi-computer architecture and affected by the initial software application, wherein the initial software application is intended to be replicated on at least one other computer in order to enable a service switch from the initial software application the a replica software application;

capturing messages issued from a client before the messages are taken into account by a network layer of the multi-computer architecture, wherein the messages are captured at the IP tables level in the context of a protocol TCP/IP;

capturing socket parameters via extended generic system calls;

modifying the socket parameters via the extended generic system calls, wherein the socket parameters that are captured and modified include at least one of local and remote ports, local and remote reference number, number of a next packet to be emitted and expected, emission and reception timer, and window size;

backing up a list of packages pending transmission, packages in transit, and packages received but not yet read by the initial software application;

transmitting the messages to the second computer over a multicast-type channel;

updating, on-the-flow, at least one clone from a plurality of clones installed on the secondary node of the initial software application, wherein the at least one clone is updated periodically, wherein the updating of the at least one clone is triggered by one or more characteristic events, and wherein each clone in the plurality of clones includes a record of the messages received by the primary node;

responsive to detecting a fault or an event affecting the first node, electing the at least one clone from the plurality of clones to be substituted for the initial software application;

switching to the at least one clone, wherein the at least one clone becomes the new primary node, and wherein the record of the messages received by the primary node are re-injected into the new primary node;

responsive to switching to the at least one clone, migrating network connections; and

supervising a state of resources necessary to operation of the initial software application.