Network fragmentation and virtual machine migration in a scalable cloud computing environment

• US 8,874,749 B1
• Filed: 02/03/2011
• Issued: 10/28/2014
• Est. Priority Date: 02/03/2010
• Status: Active Grant

• Current Assignee: Citrix Systems, Inc. (Cloud Software Group)
• Original Assignee: Citrix Systems, Inc. (Cloud Software Group)
• Inventors: Vittal, Chiradeep, Chan, Will, Huang, Alex, Liang, Sheng
• Primary Examiner(s): Srivastava, Vivek
• Assistant Examiner(s): Abedin, Normin
• Application Number:
  US13/020,802
• Time in Patent Office: 1,363 Days
• Field of Search: 709/234, 709/226, 709/224, 709/223, 709/225
• US Class Current: 709/226
• CPC Class Codes: G06F 9/5077 Logical partitioning of resources; Management or configuration of virtualized resources specific details on emulation or internal functioning of virtual machines G06F9/455
• CPC Class Codes: H04L 47/70 Admission control; Resource allocation
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers

Abstract

A scalable cloud infrastructure serves two or more customers, where each customer is associated with at least one unit of virtual resources. The virtual resources are established by apportioning physical resources in the cloud infrastructure that are partitioned into pods within one or more zones in a scalable manner. Additionally, the cloud infrastructure establishes one or more management server clusters each comprising one or more management servers. The two or more customers create a number of virtual machines within pods in a zone. Due to the scalability of the cloud infrastructure, customer virtual machines may exist in non optimal locations within the zone. A method to migrate virtual machines and defragment customer networks is devised to optimally manage network traffic and data communication in a scaled cloud infrastructure.

Claims

1. A method for defragmenting a virtual machine network of a customer, the virtual machine network comprising a plurality of virtual machines associated with the customer, the virtual machine network spanning one or more pods, wherein each pod comprises a set of physical resources in a zone, the physical resources comprising physical, storage, and network resources, the method comprising:

identifying first virtual machines of the customer in a first pod operating a first portion of the virtual machine network of the customer;

identifying additional virtual machines associated with the customer and outside the first portion of the virtual machine network;

determining a measure of network fragmentation of the customer network, the measure indicating a degree to which virtual machines of the customer are distributed across multiple pods;

determining available resources in the first pod for migrating one or more of the additional virtual machines to the first pod, and whether the measure of network fragmentation indicates an advantage to the customer network by migrating one or more of the additional virtual machines to the first pod to reduce the network fragmentation; and

based on the available resources and the measure of network fragmentation indicating the advantage to the customer network, migrating one or more of the additional virtual machines into the first portion of the customer network on the first pod,

wherein the virtual machines outside the first portion of the virtual machine network operate on one or more secondary system routing appliances in a second portion of the virtual machine network, and wherein determining the measure of network fragmentation includes considering volume of network traffic between the first and second portions of the virtual machine network.

2. A method according to claim 1, further including:

determining whether resources allocated to the one or more additional virtual machines in a second pod are less than or equal to the available resources in the first pod;

in response to determining that the allocated resources are less than or equal to the available resources in the first pod, reallocating the migrated virtual machines from the second pod to the first pod by allocating a set of corresponding physical resources in the first pod to the migrated virtual machines; and

releasing the allocated resources within the second pod.

3. The method of claim 2, wherein allocating the set of corresponding physical resources further comprises a transfer of physical storage data from the second pod to the first pod, and wherein the physical storage data includes one or more disk volumes for the migrated virtual machines.

4. The method of claim 2, wherein allocating the set of corresponding physical resources further comprises memory and processing counters, and further comprising running a migrated virtual machine from a previous processing counter.

5. The method of claim 2, wherein reallocating the migrated virtual machines further comprises routing network traffic for the migrated virtual machines from the second pod to the first pod.

6. The method of claim 1, wherein the first portion of the virtual machine network comprises a primary system routing appliance allocated to the customer.

7. The method of claim 1, wherein the measure of network fragmentation is determined from an allocation of primary and secondary system routing appliances to the virtual machine network.

8. The method of claim 1, wherein the zone comprises the first pod and a plurality of additional pods.

9. The method of claim 1, wherein the first portion of the virtual machine network and the primary routing appliance reside in the first pod, and the second portion of the virtual machine network and the one or more secondary system routing appliances reside in a second pod distinct from the first pod.

10. The method of claim 9, wherein:

the customer is a first one of a plurality of customers each having a respective customer routing domain to handle network traffic for respective virtual machine networks of the customers;

a first customer routing domain resides on a first computing server in the first pod and handles traffic for the first customer within the first portion of the virtual machine network of the first customer;

a second customer routing domain resides on a second computing server in the second pod and handles traffic for a second customer within a virtual machine network of the second customer located within the second pod;

a phantom routing domain resides on the second computing server for handling traffic in the second portion of the virtual machine network of the first customer and for communicating network traffic between the first customer routing domain and the second portion of the virtual machine network of the first customer.

11. The method of claim 1, wherein the virtual machines outside the first portion of the virtual machine network include a pair of virtual machines and another virtual machine, the pair of virtual machines using local network resources of the second portion of the customer network, and further including selecting the other virtual machine to be migrated based on its use of a greater amount of traffic of the first portion of the customer network than the pair of virtual machines.

12. A non-transitory computer readable storage medium storing instructions executable by a computer system, the instructions when executed cause the computer system to perform a method for defragmenting a virtual machine network of a customer, the virtual machine network comprising a plurality of virtual machines associated with the customer, the virtual machine network spanning one or more pods, wherein each pod comprises a set of physical resources in a zone, the physical resources comprising physical, storage, and network resources, the method including:

identifying first virtual machines of the customer in a first pod operating a first portion of the virtual machine network of the customer;

identifying additional virtual machines associated with the customer and outside the first portion of the virtual machine network;

determining a measure of network fragmentation of the customer network, the measure indicating a degree to which virtual machines of the customer are distributed across multiple pods;

determining available resources in the first pod for migrating one or more of the additional virtual machines to the first pod, and whether the measure of network fragmentation indicates an advantage to the customer network by migrating one or more of the additional virtual machines to the first pod to reduce network fragmentation; and

based on the available resources and the measure of network fragmentation indicating the advantage to the customer network, migrating one or more of the additional virtual machines into the first portion of the customer network on the first pod, wherein the virtual machines outside the first portion of the virtual machine network operate on one or more secondary system routing appliances in a second portion of the virtual machine network, and wherein determining the measure of network fragmentation includes considering volume of network traffic between the first and second portions of the virtual machine network.

13. A non-transitory computer readable storage medium according to claim 12, wherein the instructions when executed cause the one or more processors to:

determine whether resources allocated to the one or more additional virtual machines in a second pod are less than or equal to the available resources in the first pod.

14. The non-transitory computer readable storage medium of claim 13, wherein the instructions to allocate the migrated virtual machines further cause the one or more processors to route network traffic for the migrated virtual machines from the second pod to the first pod.

15. The non-transitory computer readable storage medium of claim 12, wherein the instructions further cause the one or more processors to determine the measure of network fragmentation from an allocation of primary and secondary system routing appliances to the virtual machine network.

16. The non-transitory computer readable storage medium of claim 12, wherein the first portion of the virtual machine network and the primary routing appliance reside in the first pod, and the second portion of the virtual machine network and the one or more secondary system routing appliances reside in a second pod distinct from the first pod.

17. The non-transitory computer readable storage medium of claim 12, wherein:

the customer is a first one of a plurality of customers each having a respective customer routing domain to handle network traffic for respective virtual machine networks of the customers;

a first customer routing domain resides on a first computing server in the first pod and handles traffic for the first customer within the first portion of the virtual machine network of the first customer;

a second customer routing domain resides on a second computing server in the second pod and handles traffic for a second customer within a virtual machine network of the second customer located within the second pod;

a phantom routing domain resides on the second computing server for handling traffic in the second portion of the virtual machine network of the first customer and for communicating network traffic between the first customer routing domain and the second portion of the virtual machine network of the first customer.

18. The non-transitory computer readable storage medium of claim 12, wherein the virtual machines outside the first portion of the virtual machine network include a pair of virtual machines and another virtual machine, the pair of virtual machines using local network resources of the second portion of the customer network, and further including selecting the other virtual machine to be migrated based on its use of a greater amount of traffic of the first portion of the customer network than the pair of virtual machines.