Automatic rule based grouping of compute nodes for a globally optimal cluster

Patent metadata

Field	Value
Publication number	US-10833918-B2
Application number	US-201916281128-A
Country	US
Kind code	B2
Filing date	Feb 21, 2019
Priority date	Jan 9, 2019
Publication date	Nov 10, 2020
Grant date	Nov 10, 2020

Abstract

Techniques for automatic rule based grouping of compute nodes for a global optimal cluster are disclosed. In one embodiment, a virtual infrastructure management (VIM) server may obtain a list of operating, provisioned and/or about to be provisioned compute nodes in a cluster. The VIM server may then obtain user configured and run-time network parameters of a network interface card (NIC) of each compute node in the list. Further, the VIM server may then measure round-trip times (RTTs), using a ping, between each compute node and each of remaining compute nodes in the list. Furthermore, the VIM server may then group the compute nodes in the list based on the obtained user configured and run-time network parameters and/or the measured RTTs. In addition, the VIM server may perform a high availability (HA) operation in the cluster using the grouped compute nodes.

First claim

What is claimed is:

1. A method comprising:
obtaining a list of operating, provisioned and/or about to be provisioned compute nodes in a cluster;
obtaining user configured and run-time network parameters of a network interface card (NIC) of each compute node in the list;
measuring round-trip times (RTTs), using a ping, between each compute node and each of remaining compute nodes in the list;
grouping the compute nodes in the list based on the obtained user configured and run-time network parameters and/or the measured RTTs; and
performing a high availability (HA) operation in the cluster using the grouped compute nodes.

2. The method of claim 1 , wherein the cluster comprises a stretched cluster, a multi-AZ cluster, a metro cluster, or a high availability (HA) cluster that crosses multiple areas within a local area network (LAN) or a wide area network (WAN).

3. The method of claim 1 , further comprising:
determining presence of any location devices and/or sensors in each compute node in the list; and
obtaining location related data using the location devices and/or sensors for each compute node in the list, wherein performing the HA operation in the cluster comprises:
grouping the compute nodes based on the obtained location related data, the obtained user configured and run-time network parameters, and/or the measured RTTs, and
performing the HA availability operation in the cluster using the grouped compute nodes.

4. The method of claim 1 , wherein the user configured and run-time network parameters are internet protocol (IP) routing table, address resolution protocol (ARP) table, net neighbor table, MAC address, hardware physical address, IP address, subnet mask, default gateway, domain name system (DNS) server, host name, NIC port type, port speed and/or mapping tables.

5. The method of claim 4 , wherein grouping the compute nodes in the cluster further comprises:
grouping the compute nodes in the cluster based on whether a MAC address and/or hardware physical address is in one of the ARP table and address mapping table.

6. The method of claim 1 , wherein the HA operation is a load balancing operation, fault tolerance operation and/or disaster/fault recovery operation.

7. The method of claim 1 , wherein obtaining the list of operating, provisioned and/or about to be provisioned compute nodes in the cluster comprises:
obtaining the list of operating, provisioned and/or about to be provisioned compute nodes in the cluster using a user input, an output of a provisioning tool and/or an output of a scaling tool that automatically deploys the compute nodes.

8. The method of claim 1 , wherein a minimum RTT is about 10 milliseconds.

9. The method of claim 1 , wherein measuring the RTTs, using the ping, between each compute node and each of remaining compute nodes in the list comprises:
measuring the RTTs, using a pre-determined number of pings, between each compute node and each of remaining compute nodes in the list; and
obtaining minimum RTTs from the measured RTTs obtained using the predetermined number of pings.

10. The method of claim 1 , further comprising:
configuring one or more compute nodes, that are not grouped in the list, as witness compute nodes; and
breaking ties or split-brain syndromes in the cluster during operation using the configured witness compute nodes.

11. The method of claim 1 , wherein the grouping comprises classifying, tagging, locating, and/or configuring.

12. A system comprising:
a cluster, wherein the cluster comprises one or more sites and each site comprises one or more compute nodes;
a local storage communicatively coupled to the clusters via local area networks; and
a virtual infrastructure management (VIM) server coupled to the clusters via a wide area network (WAN) is to:
obtain a list of operating, provisioned and/or about to be provisioned compute nodes in a cluster;
obtain user configured and run-time network parameters of a network interface card (NIC) of each compute node in the list;
measure round-trip times (RTTs), using a ping, between each compute node and each of remaining compute nodes in the list;
group the compute nodes in the list based on the obtained user configured and run-time network parameters and/or the measured RTTs; and
perform a high availability (HA) operation in the cluster using the grouped compute nodes.

13. The system of claim 12 , wherein the cluster comprises a stretched cluster, a multi-AZ cluster, a metro cluster, or a high availability (HA) cluster that crosses multiple areas within a local area network (LAN) or a wide area network (WAN).

14. The system of claim 12 , where in the VIM server further to:
determine presence of any location devices and/or sensors in each compute node in the list; and
obtain location related data using the location devices and/or sensors for each compute node in the list, wherein managing the HA in the cluster comprises:
group the compute nodes based on the obtained location related data, the obtained user configured and run-time network parameters, and/or the measured RTTs, and perform the HA availability operation in the cluster using the grouped compute nodes.

15. The system of claim 12 , wherein the group comprises classify, tag, locate, and/or configure.

16. The system of claim 12 , wherein the user configured and run-time network parameters are internet protocol (IP) routing table, address resolution protocol (ARP) table, net neighbor table, MAC address, hardware physical address, IP address, subnet mask, default gateway, domain name system (DNS) server, host name, NIC port type, port speed and/or mapping tables.

17. The system of claim 12 , wherein the VIM server to group the compute nodes in the cluster based on whether a MAC address and/or hardware physical address is in one of the ARP table and address mapping table.

18. The system of claim 12 , wherein the VIM server to obtain the list of operating, provisioned and/or about to be provisioned compute nodes in the cluster using a user input, an output of a provisioning tool and/or an output of a scaling tool that automatically deploys the compute nodes.

19. The system of claim 12 , wherein the HA operation is a load balancing operation, fault tolerance operation and/or disaster/fault recovery operation.

20. The system of claim 12 , wherein the VIM server obtains the list of operating, provisioned and/or about to be provisioned compute nodes in the cluster using a user input, an output of a provisioning tool and/or an output of a scaling tool that automatically deploys the compute nodes.

21. The system of claim 12 , wherein the VIM server further to:
measure RTTs, using a pre-determined number of pings, between each compute node and each of remaining compute nodes in the list; and
obtain minimum RTTs from the measured RTTs obtained using the predetermined number of pings.

22. The system of claim 12 , wherein the VIM server further to:
configure one or more compute nodes, that are not grouped in the list, as witness compute nodes; and
break ties or split-brain syndromes in the cluster during operation using the configured witness compute nodes.

23. A non-transitory machine-readable storage medium encoded with instructions that, when executed by a processor, cause the processor to:
obtain a list of operating, provisioned and/or about to be provisioned compute nodes in a cluster;
obtain user configured and run-time network parameters of a network interface car

Assignees

• Vmware Inc

Inventors

• Shang Jingliang

Classifications

• H04L43/20

  the monitoring system or the monitored elements being virtualised, abstracted or software-defined entities, e.g. SDN or NFV · CPC title

• H04L41/0893

  Assignment of logical groups to network elements · CPC title

• H04L41/40

  using virtualisation of network functions or resources, e.g. SDN or NFV entities · CPC title

• H04L41/0654 Primary

  using network fault recovery (ring fault isolation or reconfiguration in loop networks without recovery actions by a network management system H04L12/437) · CPC title

• H04L43/0864 Primary

  Round trip delays · CPC title