System and method for event-driven live migration of multi-process applications

• US 9,355,161 B1
• Filed: 11/25/2013
• Issued: 05/31/2016
• Est. Priority Date: 08/26/2005
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Open Invention Network LLC
• Inventors: Havemose, Allan
• Primary Examiner(s): Yohannes, Tesfay
• Application Number:
  US14/088,972
• Time in Patent Office: 918 Days
• Field of Search: 709/203, 709/223, 370/230, 370/352
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1458 Management of the backup or restore process
• CPC Class Codes: G06F 11/203 using migration
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• 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 16/27 Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
• CPC Class Codes: G06F 2201/805 Real-time
• CPC Class Codes: G06F 2209/542 Intercept
• CPC Class Codes: G06F 9/4856 resumption being on a different machine, e.g. task migration, virtual machine migration G06F9/5088 takes precedence
• CPC Class Codes: G06F 9/542 Event management; Broadcasting; Multicasting; Notifications
• CPC Class Codes: G06F 9/546 Message passing systems or structures, e.g. queues

Abstract

A system and method for event-driven live migration of applications between two or more servers. Primary applications runs on primary hosts and one or more replicated instances of each primary application run on one or more backup hosts. Event driven live migration is provided through a combination of process replication, logging, barrier synchronization, reliable messaging and message playback. The event driven live migration is transparent to the application and requires no modification to the application, operating system, networking stack or libraries.

Claims

1. A system, comprising:

one or more memory locations configured to store a primary application;

one or more Central Processing Units operatively connected to said one or more memory locations and configured to execute said primary application on a primary host with a host operating system and configured to send messages to one or more backup applications while said primary application is executed;

one or more backup hosts each with a host operating system and each comprising;

computer system memory comprising one or more memory locations configured to store one or more backup applications, and one or more Central Processing Units operatively connected to said computer system memory and configured to execute said one or more backup applications and configured to receive the one or more messages sent from said primary application;

one or more events configured to trigger a live migration of a primary application from a primary host to one or more backup applications on one or more backup hosts;

wherein said messages are configured to provide consistency between the primary application and said one or more backup applications; and

a barrier, wherein said barrier is configured to halt execution before or after said messages are sent to the one or more backup applications.

2. The system according to claim 1, wherein said host operating system is one of Linux, UNIX or Microsoft Windows.

3. The system according to claim 1, wherein said messages are sent over one of User Datagram Protocol (UDP), Transmission Control Protocol (TCP), UDP using multicast, or UDP using broadcast.

4. The system according to claim 1, wherein live migration of said primary application to said one or more backup hosts is performed in response to an event, wherein said event is one of operator generated live migration event, Central Processing Unit threshold event, memory threshold event, storage threshold event, Simple Network Management Protocol (SNMP) event, or script-generated event.

5. The system according to claim 1 further comprising a barrier, wherein said barrier is configured to halt execution at said barrier.

6. The system according to claim 1 further comprising a barrier, wherein said barrier is configured to halt execution while sending said messages sent to the one or more backup applications.

7. The system according to claim 1, wherein the system is configured to choose a backup host based on one of preconfigured backup, operator-chosen backup, or backup chosen based on available resources.

8. The system according to claim 1, wherein said event is generated external to said primary application.

9. The system according to claim 1, further comprising at least one of:

a barrier for the primary application;

one or more barriers for the one or more backup applications;

wherein;

said barrier for the primary application is configured to halt execution of the primary application;

said one or more barriers for the one or more backup application are configured to halt execution of the one or more backup applications;

said messages sent to the one or more backup applications are configured to generate a message for a backup to take over execution;

said backup application is configured to receive said message to take over and promote to new primary application;

said barrier for the primary application is configured to exit the primary application after sending said messages to the backup while still at the barrier; and

said barrier for the primary application is configured to halt execution at said barrier.

10. The system according to claim 9, wherein said barrier for the primary application is configured to send said message for the backup to take over execution as a replication message and said one or more backup applications are configured to receive said message to take over execution as a replication message.

11. The system according to claim 9, wherein the backup host is chosen based on one of preconfigured backup, operator-chosen backup, or backup chosen based on available resources.

12. The system according to claim 9, wherein said message for the backup to take over execution is comprised of one or more of the network Internet Protocol (IP) address of said new primary or the fully qualified network host name for said new primary.

13. The system according to claim 10, further comprising at least one of:

said backup is configured to take over execution upon receiving said message;

said barrier for the backup application is configured to exit said barrier upon receiving said message to take over execution.

14. The system according to claim 10, wherein said message to take over execution is sent and received over a non-blocking reliable protocol.

15. The system according to claim 10, wherein said message to take over execution is resent if not acknowledged.

16. The system according to claim 10, wherein duplicate messages for the backup to take over execution are rejected.

17. The system according to claim 10, wherein:

one backup is configured to be the designated backup; and

said message for the designated backup to take over execution is transmitted using Transmission Control Protocol (TCP).

18. The system according to claim 10, wherein said message to take over execution is transmitted over one of User Datagram Protocol (UDP), UDP multicast or UDP broadcast.

19. A method, comprising:

storing, at one or more memory locations, a primary application executing on a primary host with a host operating system;

generating, on said primary host, messages based on execution of said primary application;

transmitting said messages to one or more backups;

storing, at one or more additional memory locations, one or more backup applications executing on one or more backup hosts each with a corresponding host operating system;

receiving, by each one or more backup applications, said messages;

wherein said messages provide consistency between the primary application and said one or more backup applications; and

halting, by a barrier, execution before or after said messages are sent to the one or more backup applications.