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

• US 8,621,275 B1
• Filed: 12/01/2010
• Issued: 12/31/2013
• Est. Priority Date: 08/06/2010
• Status: Expired due to Fees

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Open Invention Network LLC
• Inventors: Havemose, Allan
• Primary Examiner(s): Leibovich, Yair
• Application Number:
  US12/957,593
• Time in Patent Office: 1,126 Days
• Field of Search: 714/13
• US Class Current: 714/13
• CPC Class Codes: G06F 11/1438 Restarting or rejuvenating
• CPC Class Codes: G06F 11/1448 Management of the data involved in backup or backup restore
• CPC Class Codes: G06F 11/1482 by means of middleware or OS functionality
• CPC Class Codes: G06F 11/1497 Details of time redundant execution on a single processing unit
• 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 2201/805 Real-time
• CPC Class Codes: G06F 2201/82 Solving problems relating to consistency
• 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 9/546 Message passing systems or structures, e.g. queues

Abstract

A system, method, and computer readable medium for asynchronous live migration of applications between two or more servers. The computer readable medium includes computer-executable instructions for execution by a processing system. Primary applications runs on primary hosts and one or more replicated instances of each primary application run on one or more backup hosts. Asynchronous live migration is provided through a combination of process replication, logging, barrier synchronization, checkpointing, reliable messaging and message playback. The live migration is transparent to the application and requires no modification to the application, operating system, networking stack or libraries.

Claims

1. A system for providing live migration of a primary application to one or more backup applications, the system comprising:

one or more memory locations configured to store said primary application executing on a host with a host operating system;

one or more interceptors configured to intercept calls from threads of applications to the host operating system and shared libraries for access to resources, and configured to generate replication messages based on said intercepted calls;

a barrier for said primary application that ensures that the replication messages from the primary application correspond to fully finished resource calls, by halting execution, and access to the resources and the corresponding replication messages are synchronized to the entry and exit of the interceptor;

a messaging layer for said primary application configured to transmit said replication messages to the one or more backups;

a logging facility for said messaging layer;

a checkpointing service for said primary application configured to checkpoint said primary application; and

one or more additional memory locations configured to store the one or more backup applications executing on one or more backup hosts each with a corresponding host operating system;

one or more interceptors configured to intercept calls to said one or more backup host operating systems and shared libraries;

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

a messaging layer for each one or more backup applications configured to provide ordered receipt of said replication messages;

a checkpointing service for each one or more backup applications configured to checkpoint-restore said one or more backup applications; and

wherein live migration of said primary application to said one or more backup hosts is performed in response to an event or fault.

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

3. The system according to claim 1, wherein the messaging layers are configured to transmit said messages over one of UDP, TCP, UDP using multicast, or UDP using broadcast.

4. The system according to claim 1, wherein said event is one of operator generated live migration event, CPU threshold event, memory threshold event, storage threshold event, SNMP event, or script generated event.

5. The system according to claim 1, wherein said fault is one of application crash, host crash, operating system fault, memory fault, storage fault, power supply fault, or general device fault.

6. The system according to claim 1, wherein said barrier is configured to halt execution inside said interceptors.

7. The system according to claim 1, wherein said barrier is configured to halt execution outside said intercepted resource.

8. 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 dynamically chosen backup based on available resources.

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

10. The system according to claim 1, wherein the logging facility is configured to write each message to a log set in the log on shared storage, configured to write each checkpoint to a log set in the log on shared storage, configured to create a new log set with each new checkpoint, and configure to remove old log sets when a new log set has been created.

11. The system according to claim 10, wherein the logging facility is configured to include a pending acknowledgement queue.

12. The system according to claim 10, wherein said checkpointing service for the primary application is configured to be triggered by one of a certain amount of elapsed time since the last checkpoint, a certain number of replication messages, an operator event, a resource event, or another external event.

13. The system according to claim 10, wherein a log set is comprised of a checkpoint and all replication messages between said checkpoint and the next checkpoint.

14. The system according to claim 10, wherein said logging services are configured to store the two most recent log sets on shared storage.

15. The system according to claim 1, wherein the checkpointing services for the primary application is configured to be triggered by one of a certain amount of elapsed time since the last checkpoint, a certain number of replication messages, an operator event, a resource event, or another external event.

16. The system according to claim 1, wherein the checkpointing service for the primary application is configured to place said checkpoints in replication messages and to assign said checkpoints message IDs.

17. The system according to claim 16, wherein the logging facility is configured to log said checkpoint replication messages.

18. The system according to claim 16, wherein the logging facility is configured to store checkpoint replication messages and to not transmit said checkpoint replication messages to the one or more backups.

19. The system according to claim 16, wherein the logging facility is configured to store checkpoint replication messages and to transmit said checkpoint replication messages to the one or more backups over the messaging layer.