Failure resistant distributed computing system

• US 9,519,553 B2
• Filed: 12/31/2015
• Issued: 12/13/2016
• Est. Priority Date: 12/31/2014
• Status: Active Grant

• Current Assignee: ServiceNow Incorporated
• Original Assignee: ServiceNow Incorporated
• Inventors: Chandrashekar, Sridhar, Patel, Swapnesh, Shah, Viral, Garg, Anurag, Chablani, Anjali
• Primary Examiner(s): Manoskey, Joseph D
• Application Number:
  US14/985,740
• Publication Number:
  US 20160188427A1
• Time in Patent Office: 348 Days
• Field of Search: 714/4.11, 714/4.1, 714/4.4, 714/43
• US Class Current: 1/1
• 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 11/2007 using redundant communication media
• CPC Class Codes: G06F 11/2025 using centralised failover control functionality
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: G06F 11/2048 where the redundant components share neither address space nor persistent storage
• CPC Class Codes: G06F 2201/85 Active fault masking without idle spares
• CPC Class Codes: H04L 41/0668 by dynamic selection of recovery network elements, e.g. replacement by the most appropriate element after failure
• CPC Class Codes: H04L 61/4511 using domain name system [DNS]
• CPC Class Codes: H04L 67/1034 Reaction to server failures by a load balancer
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]

Abstract

A failure resistant distributed computing system includes primary and secondary datacenters each comprising a plurality of computerized servers. A control center selects orchestrations from a predefined list and transmits the orchestrations to the datacenters. Transmitted orchestrations include less than all machine-readable actions necessary to execute the orchestrations. The datacenters execute each received orchestration by referencing a full set of actions corresponding to the received orchestration as previously stored or programmed into the computerized server and executing the referenced full set of actions. At least one of the orchestrations comprises a failover operation from the primary datacenter to the secondary datacenter. Failover shifts performance of task from a set of processing nodes of the primary datacenter to a set of processing nodes of the secondary datacenter, such tasks including managing storage accessible by one or more remote clients and running programs on behalf of remote clients.

Claims

1. A failure resistant network-based distributed computing system with a plurality of datacenters comprising:

primary and secondary datacenters, each datacenter comprising;

a computerized server comprising;

a processor configured to execute a processing node; and

a memory comprising instructions executable by the processor; and

a control center comprising;

a memory; and

a transmitter, wherein the control center is programmed to perform machine-executable operations stored in its memory to;

select orchestrations from a predefined list stored in its memory; and

transmit, using the transmitter, an identification of the selected orchestrations to at least one of the computerized servers of the primary or secondary datacenters; and

wherein;

the at least one computerized server of the primary and secondary datacenters is programmed to perform machine-executable operations to, responsive to receiving identification of one of the selected orchestrations from the control center, transmit, to the control center, a request for an updated list of machine-executable actions necessary to execute an identified orchestration, and execute the identified orchestration using its processor by referencing a set of actions corresponding to the identified orchestration as previously stored or programmed into the computerized server and executing the set of actions on the server processor modified by the updated list;

at least one of the machine-executable actions is to direct at least one other computerized server to execute prescribed tasks on its processor; and

the predefined list of orchestrations comprises at least one machine-executable orchestration to conduct a failover operation from the primary datacenter to the secondary datacenter, the failover operation comprising shifting performance of tasks from at least one processing node of the primary datacenter to at least one processing node of the secondary datacenter, the tasks comprising;

managing storage accessible by one or more clients located remotely from the datacenters; and

running programs of machine-implemented operations on behalf of clients remotely located from the datacenters.

2. The system of claim 1, wherein at least one of the machine-executable actions corresponding to the identified orchestration comprises a given computerized server transferring control of execution on its processor of other actions corresponding to the identified orchestration to a different computerized server processor than the given computerized server processor.

3. The system of claim 2, wherein the action of the given computerized server transferring control comprises executing at least one of the actions on the processor corresponding to the identified orchestration on the different computerized server.

4. The system of claim 1, wherein at least one given machine-executable action corresponding to an orchestration comprises transferring control of execution of future actions corresponding to the orchestration to a different computerized server than is performing the given machine-executable action.

5. The system of claim 1, wherein:

the request includes a first version of a set of actions necessary to execute the received orchestration; and

the control center further comprises instructions executable by its processor to perform operations comprising;

comparing the first version against a second version of a set of actions necessary to execute the received orchestration maintained by the control center; and

responsive to detecting one or more differences between the first version and second version, transmitting data representing changes between the first version and the second version to the at least one computerized server.

7. The system of claim 1, where at least a first one of the orchestrations in the predefined list corresponds to one or more machine-executable actions to override one or more of a full set of machine-executable actions stored or programmed into one or more of the computerized servers.

8. The system of claim 1, wherein the at least one computerized server further comprises instructions executable by its processor such that the set of predefined machine-executable actions are stored or programmed into the memory of the computerized servers by:

storage accessible by the computerized servers;

incorporation into circuitry of the computerized servers;

or

incorporation into code executable by the servers.

9. The system of claim 1, wherein one of the computerized servers comprises instructions executable by its processor to provide database services.

10. The system of claim 1, wherein one of the computerized servers comprises instructions executable by its processor to execute a plurality of virtual machine instantiations, each virtual machine instantiation configured to execute machine-executable actions on behalf of a remote client.

11. The system of claim 1, wherein each computerized server of the primary and secondary datacenters comprises a mailbox, and comprises instructions executable by its processor to execute commands received at the mailbox of the computerized server.

12. The system of claim 1, wherein each computerized server of the primary and secondary datacenters comprises a mailbox, and comprises instructions executable by its processor to ignore incoming messages arriving at the mailbox if a task specified by the message is already in progress at the computerized server.

13. The system of claim 1, wherein one of the orchestrations stored in the memory corresponds to a set of machine-executable actions comprising one computerized server commanding another computerized server to perform a task.

14. The system of claim 1, wherein a first orchestration from the predefined list stored in the memory comprises actions executable by one of the computerized servers of the secondary datacenter to begin operating the secondary datacenter in substitution for the primary datacenter.

15. The system of claim 14, wherein the actions stored in the memory to begin operating the secondary datacenter in substitution for the primary datacenter comprise:

a stop processing action that stops processing nodes provided by the computerized servers of the primary and secondary datacenters;

a read-only mode action that, for all databases provided by computerized servers of the primary datacenter, places the databases in a read-only mode;

a read-and-write mode action that, for all databases provided by computerized servers of the secondary datacenter, places the databases in a read-and-write mode;

a re-routing action that re-routes connections of the processing nodes of computerized servers of the primary datacenter to processing nodes of computerized servers of the secondary datacenter; and

a restarting action that restarts processing nodes provided by computerized servers of the primary and secondary datacenters.

16. The system of claim 1, wherein:

the processing nodes comprise a virtual machine instance to perform tasks including running applications and operating databases, and

each of the computerized servers further comprises an agent to manage the processing nodes and control prescribed operations of the computerized server apart from the processing nodes.

6. A failure resistant network-based distributed computing system with a plurality of datacenter, comprising:

primary and secondary datacenters, each datacenter comprising;

a plurality of computerized servers, each of the computerized servers comprising;

a processor;

a communications port connected to a network;

a memory comprising instructions executable by the processor; and

a messaging queue connected via the communications port with the computerized servers of the datacenter;

wherein;

the processor is configured to execute a processing node; and

the messaging queues of the primary and secondary datacenters are communicatively interconnected via their respective communication ports by one or more links;

the system further comprising;

a control center comprising;

one or more digital data processing machines;

a communications port;

a memory; and

a transmitter that communicates via signals sent over its communications port coupled to the at least one messaging queue of each datacenter, wherein the control center is programmed to perform machine-executable operations stored in its memory to;

select orchestrations from a predefined list stored in its memory; and

transmit, using the transmitter, an identification of the selected orchestrations to a server of the computerized servers of the primary or secondary datacenters via a respective one of the messaging queues; and

wherein;

each of the computerized servers of the primary and secondary datacenters is programmed to perform machine-executable operations to, responsive to receiving identification of one of the selected orchestrations from the control center via one of the messaging queues, execute the identified orchestration using its processor by referencing a full set of actions corresponding to the received orchestration as previously stored or programmed into the computerized server and executing the referenced full set of actions on the server processor; and

at least one of the machine-executable actions is to direct at least one other computerized server to execute prescribed tasks on its processor;

the predefined list of orchestrations comprises at least one machine-executable orchestration to conduct a failover operation from the primary datacenter to the secondary datacenter, the failover operation comprising shifting performance of tasks from a set of processing nodes of the primary datacenter to a set of processing nodes of the secondary datacenter, the tasks comprising;

managing storage accessible by one or more clients located remotely from the datacenters; and

running programs of machine-implemented operations on behalf of clients remotely located from the datacenters; and

at least one of the computerized servers comprises instructions executable by its processor to perform operations comprising;

from the control center, receiving a differences list corresponding to a given orchestration; and

performing the given orchestration by executing an amended set of predefined machine-executable actions, the amended set of predefined machine-executable actions comprising the full set of predefined machine-executable actions necessary to execute the given orchestration as stored or programmed into the memory of the computerized server and further as amended according to the differences list.

17. A computer-implemented method of operating a failure resistant distributed computing system comprising a primary datacenter and a secondary datacenter, each datacenter comprising a computerized server comprising a processor configured to execute a processing node, wherein the primary datacenter and the secondary datacenter are communicatively interconnected, the system further comprising a control center, the method comprising machine-executed operations of:

selecting with the control center orchestrations from a predefined list stored in the control center;

transmitting via the control center an identification of the selected orchestrations to the computerized server of at last one of the primary datacenter or the secondary datacenter; and

performing operations by the computerized servers of the primary datacenter and the secondary datacenters comprising;

receiving identification of one of the selected orchestrations from the control center; and

responding to the receiving identification by executing the identified orchestration by referencing a set of actions corresponding to the received orchestration as previously stored or programmed into the computerized server and executing the referenced set of actions;

wherein;

at least one of the referenced set of actions comprises directing at least one other computerized server to execute prescribed tasks; and

the predefined list of orchestrations comprises at least one machine-executable orchestration to conduct a failover operation from the primary datacenter to the secondary datacenter;

the failover operation comprises;

executing pre-validation tests at the primary datacenter;

stopping the failover operation if any of the pre-validation tests fail; and

unless any of the pre-validation tests fail;

shifting performance of the tasks from a set of processing nodes of the primary datacenter to a set of processing nodes of the secondary datacenter, the tasks comprising;

managing storage accessible by one or more clients located remotely from the datacenters; and

running programs of machine-implemented operations on behalf of clients remotely located from the datacenters.

18. The method of claim 17, further comprising:

before executing the identified orchestration, transmitting to the control center a request for an updated list of actions necessary to execute the identified orchestration.

19. The method of claim 17, wherein the failover operation further comprises, before the shifting:

stopping any of the set of processing nodes being executed by the computerized server of the primary datacenter;

stopping any of the set of processing nodes being executed by the computerized server of the secondary datacenter;

placing databases of the primary datacenter in read-only mode; and

placing databases of the secondary datacenter in read-write mode;

wherein the shifting comprises;

re-routing connections of the set of processing nodes of the primary datacenter to the set of processing nodes of the secondary datacenter; and

wherein the failover operation further comprises, after the shifting;

restarting any stopped processing node being executed by the computerized servers of the primary datacenter and the secondary datacenter.

20. A failure resistant network-based distributed computing system with a plurality of datacenters, comprising:

a primary datacenter and a secondary datacenter, each datacenter comprising a computerized server programmed to perform machine-executable operations to, responsive to receiving identification of a selected orchestrations from a control center, execute the identified orchestration using its a processor of the computerized server by referencing a full set of actions corresponding to the identified orchestration as previously stored or programmed into the computerized server and executing the full set of actions on the processor of the computerized server; and

at least one of the machine-executable operations is to direct at least one other computerized server to execute prescribed tasks on its processor; and

a predefined list of orchestrations comprising;

a machine-executable orchestration to switch DNS entries;

a machine-executable orchestration to gather topology information;

a machine-executable orchestration to conduct a failover operation from the primary datacenter to the secondary datacenter, the failover operation comprising shifting performance of tasks from a set of processing nodes of the primary datacenter to a set of processing nodes of the secondary datacenter, the tasks comprising;

managing storage accessible by one or more clients located remotely from the datacenters; and

running programs of machine-implemented operations on behalf of clients remotely located from the datacenters; and

a machine-executable orchestration to update a CMDB state.

21. The failure resistant network-based distributed computing system of claim 20, wherein the machine-executable operations include operations programmed to, before executing the identified orchestration, receive a differences list corresponding to the received orchestration such that the full set of actions to be executed is amended according to the differences list.