System and method for elastic scaling using a container-based platform

• US 9,462,427 B2
• Filed: 01/13/2016
• Issued: 10/04/2016
• Est. Priority Date: 01/14/2015
• Status: Active Grant

• Current Assignee: Kodiak Networks Inc (Motorola Solutions, Inc.)
• Original Assignee: Kodiak Networks Inc (Motorola Solutions, Inc.)
• Inventors: Patel, Krishnakant M., Kandula, Ramu, Vempati, Brahmananda R., Negalaguli, Harisha Mahabaleshwara, Chandana, Pratap
• Primary Examiner(s): SHARMA, SUJATHA R
• Application Number:
  US14/994,757
• Publication Number:
  US 20160205518A1
• Time in Patent Office: 265 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: H04L 41/0893 Assignment of logical groups to network elements
• CPC Class Codes: H04L 41/0895 Configuration of virtualised networks or elements, e.g. virtualised network function or OpenFlow elements
• CPC Class Codes: H04L 41/0897 by horizontal or vertical scaling of resources, or by migrating entities, e.g. virtual resources or entities
• CPC Class Codes: H04L 41/12 Discovery or management of network topologies
• CPC Class Codes: H04L 41/5041 characterised by the time relationship between creation and deployment of a service
• CPC Class Codes: H04L 65/1045 Proxies, e.g. for session initiation protocol [SIP]
• CPC Class Codes: H04L 65/1069 Session establishment or de-establishment
• CPC Class Codes: H04L 65/1073 Registration or de-registration
• CPC Class Codes: H04L 65/4061 Push-to services, e.g. push-to-talk or push-to-video
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/1017 based on a round robin mechanism
• CPC Class Codes: H04L 67/1031 Controlling of the operation of servers by a load balancer, e.g. adding or removing servers that serve requests
• CPC Class Codes: H04L 67/51 Discovery or management thereof, e.g. service location protocol [SLP] or web services
• CPC Class Codes: H04M 7/006 Networks other than PSTN/ISDN providing telephone service, e.g. Voice over Internet Protocol (VoIP) , including next generation networks with a packet-switched transport layer H04L65/00 takes precedence; aspects not specific to the type of network H04M3/00; special services in those networks H04M3/42
• CPC Class Codes: H04W 4/08 User group management
• CPC Class Codes: H04W 4/10 Push-to-Talk [PTT] or Push-On-Call services

Abstract

An embodiment method includes triggering, by a service orchestrator, creation of one or more container instances for a first service cluster. The method further includes creating, by a container manager, the one or more container instances and mapping the one or more container instances of the first service cluster to one or more first virtual machines belonging to a first virtual machine server group in accordance with a platform profile of the first virtual machine server group and the first service provided by the first service cluster. The method further includes mapping, by a virtual machine manager, the one or more first virtual machines to one or more first host virtual machines of a cloud network in accordance with the platform profile of the first virtual machine server group.

Claims

1. A method comprising:

triggering, by a service orchestrator hosted on a processor, creation of one or more container instances for a first service cluster, wherein the first service cluster provides a first service for a telecommunications services platform;

creating, by a container manager hosted on a processor, the one or more container instances;

mapping, by the container manager, the one or more container instances of the first service cluster to one or more first virtual machines belonging to a first virtual machine server group in accordance with a platform profile of the first virtual machine server group and the first service provided by the first service cluster;

mapping, by a virtual machine manager hosted on a processor, the one or more first virtual machines to one or more first host virtual machines of a cloud network in accordance with the platform profile of the first virtual machine server group; and

deploying the one or more first host virtual machines on one or more host processors.

2. The method of claim 1, wherein the platform profile comprises computer processing unit (CPU) parameters, network interface parameters, block storage parameters, random access memory parameters, or a combination thereof provided by each of the one or more first virtual machines belonging to the first virtual machine server group.

3. The method of claim 1 further comprising:

registering the one or more container instances with a service registrar; and

adding, by the service registrar, internet protocol (IP) addresses for interfaces of the one or more container instances with a domain name system (DNS), wherein the IP addresses for the interfaces of the one or more container instances are each in accordance with a fully qualified domain name (FQDN) of the first service, and wherein the interfaces of the one or more container instances are discoverable by other components of the telecommunications services platform by performing a DNS search on the FQDN.

4. The method of claim 3 further comprising restricting discovery of the interfaces of the one or more container instances to components of the telecommunications services platform deployed on processors located at a same deployment site as the one or more host processors.

5. The method of claim 3, wherein the other components of the telecommunications services platform are deployed across all deployment sites of the telecommunications services platform.

6. The method of claim 1 further comprising ensuring, by the container manager, at least a minimum number of virtual machines are available in the first virtual machine server group for the first service cluster.

7. The method of claim 1, wherein mapping the one or more container instances to the one or more first virtual machines comprises mapping the one or more container instances to the one or more first virtual machines in accordance with an anti-affinity policy.

8. The method of claim 7, wherein the anti-affinity policy comprises:

mapping a first one of the one or more container instances to a first one of the one or more first virtual machines; and

mapping a second container instance of the first service cluster to the first one of the one or more first virtual machines only when all virtual machines belonging to the first virtual machine server group host at least as many container instances for the first service cluster as the first one of the one or more first virtual machines.

9. The method of claim 1 further comprising:

bundling a first one of the one or more container instances with a second container instance, wherein the second container instance belongs to a different service cluster than the first service cluster, wherein the different service cluster provides a second service for the telecommunications services platform different than the first service; and

mapping the second container instance to a same one of the one or more first virtual machines as the first one of the one or more container instances.

10. The method of claim 1 further comprising:

triggering, by the service orchestrator, a ramp down of container instances in accordance with one or more service metrics of the telecommunications services platform; and

removing, by the container manager, one or more second container instances in accordance with a ramp down trigger received from the service orchestrator.

11. The method of claim 10 further comprising forcing, by the container manager, the one or more second container instances into a drain state before removing the one or more second container instances and after receiving the ramp down trigger.

12. The method of claim 11, wherein a container completes ongoing transactions or transfers sessions and pending transactions to a different container when the container is in the drain state, and wherein a load balancer stops sending new transactions to the container when the container is in the drain state.

13. The method of claim 11, further comprising forcing, by the container manager, a container instance into the drain state when the container manager updates the container instance.

14. The method of claim 10 further comprising

setting, by the container manager, a ramp down guard timer when the one or second container instances are removed;

and

not removing, by the container manager, any additional container instances in accordance with another ramp down trigger until the ramp down guard timer expires.

15. The method of claim 1 further comprising:

determining, by the container manager, when a first one of the one or more first virtual machines is overloaded due to load skew;

removing, by the container manager, a first one of the one or more container instances on the first one of the one or more first virtual machines; and

creating, by the container manager, a replacement container instance for the first service cluster on a different virtual machine belonging to the first virtual machine server group to replace the first one of the one or more container instances.

16. The method of claim 1 further comprising grouping virtual machines in the telecommunications services platform into virtual machine server groups in accordance with platform parameters provided by each of the virtual machines, wherein each virtual machine server group has a different platform profile.

17. The method of claim 16, wherein each service provided by the telecommunications services platform is mapped to a different virtual machine server group.

18. The method of claim 1, wherein mapping the one or more virtual machines to one or more host virtual machines of a cloud network in accordance with the platform profile of the first virtual machine server group comprises mapping the one or more virtual machines to one or more host virtual machines providing platform characteristics best matching the platform profile of the first virtual machine server group compared to other types of host virtual machines provided by the cloud network.

19. A method comprising:

triggering, by a processor, creation of one or more container instances for a first service cluster, wherein the first service cluster is one of a plurality of service clusters in a push-to-talk (PTT) platform, and wherein each of the plurality of service clusters provides a different function for the PTT platform;

creating, by a processor, the one or more container instances;

mapping, by a processor, the one or more container instances of the first service cluster to one or more first virtual machines belonging to a first virtual machine server group in accordance with a platform profile of the first virtual machine server group and a first PTT function provided by the first service cluster;

mapping, by a virtual machine manager hosted on a processor, the one or more virtual machines to one or more first host virtual machines of a cloud network in accordance with the platform profile of the first virtual machine server group, wherein the cloud network is deployed independently from the PTT platform; and

operating the one or more first host virtual machines on one or more host processors to provide the first PTT function for the PTT platform.

20. The method of claim 19 further comprising directing traffic originating from a PTT client on a user equipment (UE) to a deployment site in accordance with a geographic proximity of the deployment site to the UE.

21. The method of claim 19 further comprising directing traffic originating from a PTT client on a user equipment (UE) to a deployment site in accordance with a weighted round-robin load balancing policy.

22. The method of claim 19 further comprising:

receiving a registration request from a PTT client on a user equipment (UE); and

selecting a deployment site for the registration request, wherein all subsequent session initial protocol (SIP) service requests for the PTT client are directed to the deployment site.

23. The method of claim 19 further comprising supporting, by a load balancer hosted on a processor, container instance-stickiness, for session based workloads of a PTT client on a user equipment (UE).

24. The method of claim 19, wherein triggering the creation of one or more container instances for a first service cluster is in accordance with one or more service metrics, wherein the one or more service metrics comprises number of push-to-talk (PTT) pre-established sessions, PTT call setup rate, PTT call leg setup rate, number of concurrently active PTT calls, number of concurrently active PTT call legs, number of media codec instances in active use, or a combination thereof.

25. The method of claim 19 further comprising:

maintaining a first signaling path between a PTT client on a user equipment (UE) and a first deployment site of the PTT platform; and

maintaining a second signaling path between the PTT client and a second deployment site of the PTT platform, wherein the first signaling path and the second signaling path are maintained simultaneously, and wherein the first deployment site is at a different geographic location than the second deployment site.

26. The method of claim 25 further comprising transferring active sessions of the PTT client using the second signaling path when the first deployment site fails.

27. The method of claim 19 further comprising:

detecting, by a service discovery mechanism, a first deployment site of the PTT platform is partitioned from one or more second deployment sites of the PTT platform;

determining, by the service discovery mechanism, one or more third deployment sites connected the first deployment site after the first deployment site is partitioned; and

forcing, by the service discovery mechanism, the first deployment site into a dormant state when the one or more third deployment sites and the first deployment site do not account for a majority of deployment sites in the PTT platform.

28. The method of claim 27 further comprising transferring traffic from the first deployment site to the one or more second deployment sites when the first deployment site is in the dormant state.

29. A telecommunications services platform comprising:

one or more processors; and

a non-transitory computer readable storage medium storing programming for execution by the one or more processors, the programming including instructions to;

trigger, by a service orchestrator, creation of one or more container instances for a service cluster, wherein the service cluster provides a function for the telecommunications services platform;

create, by a container manager, the one or more container instances;

map, by the container manager, the one or more container instances of the service cluster to one or more virtual machines belonging to a virtual machine server group in accordance with a platform profile of the virtual machine server group and the function provided by the service cluster;

map, by a virtual machine manager hosted on a processor, the one or more virtual machines to one or more host virtual machines of a cloud network in accordance with the platform profile of the virtual machine server group; and

operate the one or more host virtual machines to provide the function for the telecommunications services platform.

30. The telecommunications services platform of claim 29, wherein each of the one or more host virtual machines comprise a plurality of first network ports, wherein each of the plurality of first network ports provides a connections to a different type of network traffic.

31. The telecommunications services platform of claim 30, wherein each of the one or more container instances comprises a plurality of second network ports, and wherein each of the one or more host virtual machines comprises one or more virtual network switches to route traffic between the plurality of first network ports and the plurality of second network ports.