Provisioning a cluster of distributed computing platform based on placement strategy

• US 9,268,590 B2
• Filed: 02/29/2012
• Issued: 02/23/2016
• Est. Priority Date: 02/29/2012
• Status: Active Grant

• Current Assignee: Vmware LLC (Broadcom, Inc.)
• Original Assignee: VMware, Inc. (Broadcom, Inc.)
• Inventors: Du, Junping, He, Ying, Wan, Da, Xiao, Jun
• Primary Examiner(s): Rashid, Wissam
• Application Number:
  US13/407,895
• Publication Number:
  US 20130227558A1
• Time in Patent Office: 1,455 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• 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 43/04 Processing captured monitoring data, e.g. for logfile generation
• CPC Class Codes: H04L 67/52 specially adapted for the location of the user terminal

Abstract

Embodiments perform automated provisioning of a cluster for a distributed computing platform. Target host computing devices are selected from a plurality of host computing devices based on configuration information, such as a desired cluster size, a data set, code for processing the data set and, optionally, a placement strategy. One or more virtual machines (VMs) are instantiated on each target host computing device. Each VM is configured to access a virtual disk that is preconfigured with code for executing functionality of the distributed computing platform and serves as a node of the cluster. The data set is stored in a distributed file system accessible by at least a subset of the VMs. The code for processing the data set is provided to at least a subset of the VMs, and execution of the code is initiated to obtain processing results.

Claims

1. A method for provisioning a cluster for a distributed computing platform, the method comprising:

receiving configuration information for the cluster, wherein the configuration information comprises at least a cluster size, a data set, and code for processing the data set;

selecting a plurality of target host computing devices from a plurality of host computing devices based on the configuration information;

instantiating, based on the cluster size, at least one virtual machine (VM) on each of the target host computing devices to serve as a node of the cluster, wherein each instantiated VM is configured to access a virtual disk that is based on a VM template in a set of VM templates and the at least one VM is preconfigured with distributed software computing code for executing functionality of the distributed computing platform based on the respective VM template;

receiving physical location information for a plurality of racks in which each instantiated VM is located, wherein a rack includes multiple host computing devices of the plurality of target host computing devices;

providing the physical location information to an instantiated VM, wherein the instantiated VM uses the physical location information to determine where to store the data set in a distributed file system accessible by at least a subset of the VMs based on a placement strategy for processing of the data set, wherein the placement strategy can be a placement strategy for operational robustness or operational efficiency or a combination of operational robustness and operational efficiency, wherein, given a placement strategy for operational robustness, data for the data set is stored in different racks using the physical location information, wherein, given a placement strategy for operational efficiency, replica data of the data for the data set is stored in a same location as the data for the data set, and wherein, given a placement strategy for a combination of operational robustness and operational efficiency, a first replica of the data for the data set is placed at a different location from that of original data for the data set and a second replica of the data for the data set is placed at the same location as that of the data for the data set and the distributed file system is accessed by the distributed computing platform during processing of the data set;

providing the code for processing the data set to at least a subset of the VMs; and

initiating execution of the code for processing the data set on the at least subset of VMs to obtain data processing results, wherein the at least subset of VMs use the distributed software computing code to execute the code for processing the data set.

2. The method of claim 1, wherein at least one of the instantiated VMs is configured to utilize local storage in the target host computing device on which the VM has been instantiated, and the local storage supports the distributed file system.

3. The method of claim 2, wherein different portions of the data set is stored in different local storages corresponding to target host computing devices of the instantiated VMs.

4. The method of claim 1, further comprising the step of selecting the VM template that serves as a basis for virtual disks based on the configuration information.

5. The method of claim 4, wherein at least one virtual disk of an instantiated VM is a linked clone of the VM template.

6. The method of claim 1, further comprising the step of enabling a user to maintain a life cycle of the cluster for future processing of data.

7. The method of claim 1, wherein the step of selecting the plurality of target host computing devices is based on the placement strategy and a location of each of the target host computing devices, wherein the location includes one or more of the following:

a physical location and a network location.

8. The method of claim 1, further comprising:

generating a script for use by each instantiated VM to identify the physical location information for each instantiated VM; and

providing the script to each instantiated VM, wherein each instantiated VM uses the script to provide respective location information for each instantiated VM.

9. One or more non-transitory computer-readable storage media including computer-executable instructions that, when executed by a computer processor, cause the computer processor to provision a cluster of a distributed computing platform having a plurality of virtual machines (VMs) by:

receiving configuration information for the cluster, wherein the configuration information comprises at least a cluster size, a data set, and code for processing the data set;

selecting a plurality of target host computing devices from a plurality of host computing devices based on the configuration information;

instantiating, based on the cluster size, at least one VM on each of the target host computing devices to serve as a node of the cluster, wherein each instantiated VM is configured to access a virtual disk that is based on a VM template in a set of VM templates and the at least one VM is preconfigured with distributed software computing code for executing functionality of the distributed computing platform based on the respective VM template

receiving physical location information for a plurality of racks in which each instantiated VM is located, wherein a rack includes multiple host computing devices of the plurality of target host computing devices;

providing the physical location information to an instantiated VM, wherein the instantiated VM uses the physical location information to determine where to store the data set in a distributed file system accessible by at least a subset of the VMs based on a placement strategy for processing of the data set, wherein the placement strategy can be a placement strategy for operational robustness or operational efficiency or a combination of operational robustness and operational efficiency, wherein, given a placement strategy for operational robustness, data for the data set is stored in different racks using the physical location information, wherein, given a placement strategy for operational efficiency replica data of the data for the data set is stored in a same location as the data for the data set, and wherein, given a placement strategy for a combination of operational robustness and operational efficiency, a first replica of the data for the data set is placed at a different location from that of original data for the data set and a second replica of the data for the data set is placed at the same location as that of the data for the data set and the distributed file system is accessed by the distributed computing platform during processing of the data set;

providing the code for processing the data set to at least a subset of the VMs; and

initiating execution of the code for processing the data set on the at least subset of VMs to obtain data processing results, wherein the at least subset of VMs use the distributed software computing code to execute the code for processing the data set.

10. The non-transitory computer-readable storage media of claim 9, wherein at least one of the instantiated VMs is configured to utilize a local storage in the target host computing device on which the VM has been instantiated, and the local storage supports the distributed file system.

11. The non-transitory computer-readable storage media of claim 10, wherein the computer-executable instructions cause the computer processor to store different portions of the data set in different local storages corresponding to target host computing devices of the instantiated VMs.

12. The non-transitory computer-readable storage media of claim 9, wherein the computer-executable instructions further cause the computer processor to select the VM template that serves as a basis for virtual disks based on the configuration information.

13. The non-transitory computer-readable storage media of claim 12, wherein at least one virtual disk of an instantiated VM is a linked clone of the VM template.

14. The non-transitory computer-readable storage media of claim 9, wherein the computer-executable instructions further cause the computer processor to enable a user to maintain a life cycle of the cluster for future processing of data.

15. The non-transitory computer-readable storage media of claim 9, wherein the step of selecting the target host computing devices is based on the placement strategy and a location of each of the target host computing devices, wherein the location includes one or more of the following:

a physical location and a network location.

16. A system for provisioning a cluster of a distributed computing platform, the system comprising:

a plurality of host computing devices; and

a management device coupled in communication with the host computing devices and configured to;

receive configuration information for the cluster, wherein the configuration information comprises at least a cluster size, a data set, and code for processing the data set;

select a plurality of target host computing devices from the plurality of host computing devices based on the configuration information;

instantiate, based on the cluster size, at least one virtual machine (VM) on each of the target host computing devices to serve as a node of the cluster, wherein each instantiated VM is configured to access a virtual disk that is based on a VM template in the set of VM templates and the at least one VM is preconfigured with distributed software computing code for executing functionality of the distributed computing platform based on the respective VM template;

receive physical location information for a plurality of racks in which each instantiated VM is located, wherein a rack includes multiple host computing devices of the plurality of target host computing devices;

provide the physical location information to an instantiated VM, wherein the instantiated VM uses the physical location information to determine where to store the data set in a distributed file system accessible by at least a subset of the VMs based on a placement strategy for processing of the data set, wherein the placement strategy can be a placement strategy for operational robustness or operational efficiency or a combination of operational robustness and operational efficiency, wherein, given a placement strategy for operational robustness, data for the data set is stored in different racks using the physical location information, wherein, given a placement strategy for operational efficiency, replica data of the data for the data set is stored in a same location as the data for the data set, and wherein, given a placement strategy for a combination of operational robustness and operational efficiency, a first replica of the data for the data set is placed at a different location from that of original data for the data set and a second replica of the data for the data set is placed at the same location as that of the data for the data set and the distributed file system is accessed by the distributed computing platform during processing of the data set;

provide the code for processing the data set to at least a subset of the VMs; and

initiate execution of the code for processing the data set on the at least subset of VMs to obtain data processing results, wherein the at least subset of VMs use the distributed software computing code to execute the code for processing the data set.

17. The system of claim 16, wherein at least one of the instantiated VMs is configured to utilize a local storage in the target host computing device on which the VM has been instantiated, and the local storage supports the distributed file system.

18. The system of claim 17, wherein the management device is configured to store different portions of the data set in different local storages corresponding to target host computing devices of the instantiated VMs.

19. The system of claim 16, wherein the management device is further configured to select the VM template that serves as a basis for virtual disks based on the configuration information.

20. The system of claim 19, wherein the management device is further configured to enable a user to maintain a life cycle of the cluster.

21. The system of claim 16, wherein the management device is configured to select the target host computing devices based on the placement strategy and a location of each of the target host computing devices, wherein the location includes one or more of the following:

a physical location and a network location.