System and method for structuring self-provisioning workloads deployed in virtualized data centers
View Patent ↗The system and method for structuring self-provisioning workloads deployed in virtualized data centers described herein may provide a scalable architecture that can inject intelligence and embed policies into managed workloads to provision and tune resources allocated to the managed workloads, thereby enhancing workload portability across various cloud and virtualized data centers. In particular, the self-provisioning workloads may have a packaged software stack that includes resource utilization instrumentation to collect utilization metrics from physical resources that a virtualization host allocates to the workload, a resource management policy engine to communicate with the virtualization host to effect tuning the physical resources allocated to the workload, and a mapping that the resource management policy engine references to request tuning the physical resources allocated to the workload from a management domain associated with the virtualization host.
1. A system for structuring self-provisioning workloads deployed in virtualized data centers, comprising:
a data center manager, wherein the data center manager includes one or more processors configured to deploy a workload to one or more virtualization hosts that allocate one or more physical hardware resources to the workload;
resource utilization instrumentation structured within the workload, wherein the workload structures the resource utilization instrumentation to collect utilization metrics from the one or more physical hardware resources that the one or more virtualization hosts allocate to the workload, wherein the resource utilization instrumentation includes a packaged module that a kernel associated with the workload loads into an operating system that hosts the workload;
a resource management policy engine further structured within the workload, wherein the workload structures the resource management policy engine to communicate with the one or more virtualization hosts to request tuning the physical hardware resources allocated to the workload in response to determining that the utilization metrics collected with the resource utilization instrumentation violates one or more resource utilization policies that the resource management policy engine defines to manage the workload; and
one or more virtual machine configuration files further structured within the workload to declare one or more limits or ranges for the physical hardware resources that the workload requires, wherein the resource management policy engine determines that the utilization metrics collected with the resource utilization instrumentation violates the one or more resource utilization policies in response to the collected utilization metrics failing to comply with the one or more limits or ranges declared in the virtual machine configuration files.
2. The system of claim 1 , wherein the utilization metrics with the resource utilization instrumentation include one or more of processing utilization metrics, memory utilization metrics, input/output bandwidth utilization metrics, or storage utilization metrics.
3. The system of claim 1 , wherein the resource management policy engine averages the collected utilization metrics over a predetermined time period to determine whether the collected utilization metrics comply with the declared limits or ranges.
4. The system of claim 1 , wherein the resource management policy engine computes the collected utilization metrics instantaneously to determine whether the collected utilization metrics comply with the declared limits or ranges.
5. The system of claim 1 , wherein the resource management policy engine further defines one or more actions that the resource management policy engine initiates to remediate specific classes of the physical hardware resources allocated to the workload violating the one or more resource utilization policies.
6. The system of claim 1 , further comprising a virtualization host specific mapping further structured within the workload, wherein the workload further structures the resource management policy engine to reference the virtualization host specific mapping to request tuning the physical hardware resources allocated to the workload from one or more management domains associated with the one or more virtualization hosts.
7. The system of claim 6 , wherein the one or more management domains associated with the one or more virtualization hosts tune the physical hardware resources that the one or more virtualization hosts allocate to the workload in response to the one or more virtualization hosts having available physical hardware resources sufficient to comply with the one or more limits or ranges declared in the one or more virtual machine configuration files.
8. A method for structuring self-provisioning workloads deployed in virtualized data centers, comprising:
deploying, by a data center manager that includes one or more processors, a workload to one or more virtualization hosts, wherein the one or more virtualization hosts allocate one or more physical hardware resources to the workload;
collecting, by resource utilization instrumentation structured within the workload, utilization metrics from the one or more physical hardware resources that the one or more virtualization hosts allocate to the workload, wherein the resource utilization instrumentation includes a packaged module that a kernel associated with the workload loads into an operating system that hosts the workload;
communicating with the one or more virtualization hosts to request tuning the physical hardware resources allocated to the workload, wherein a resource management policy engine further structured within the workload communicates with the one or more virtualization hosts to request tuning the physical hardware resources allocated to the workload in response to determining that the utilization metrics collected with the resource utilization instrumentation violates one or more resource utilization policies defined to manage the workload; and
declaring one or more limits or ranges for the physical hardware resources that the workload requires within one or more virtual machine configuration files further structured within the workload, wherein the resource management policy engine determines that the utilization metrics collected with the resource utilization instrumentation violates the one or more resource utilization policies in response to the collected utilization metrics failing to comply with the one or more limits or ranges declared in the virtual machine configuration files.
9. The method of claim 8 , wherein the utilization metrics collected with the resource utilization instrumentation include one or more of processing utilization metrics, memory utilization metrics, input/output bandwidth utilization metrics, or storage utilization metrics.
10. The method of claim 8 , further comprising loading the packaged module that includes the resource utilization instrumentation into the operating system that hosts the workload.
11. The method of claim 8 , wherein the resource management policy engine averages the collected utilization metrics over a predetermined time period to determine whether the collected utilization metrics comply with the declared limits or ranges.
12. The method of claim 8 , wherein the resource management policy engine computes the collected utilization metrics instantaneously to determine whether the collected utilization metrics comply with the declared limits or ranges.
13. The method of claim 8 , wherein the resource management policy engine initiates one or more actions to remediate specific classes of the physical hardware resources allocated to the workload in response to the collected utilization metrics violating the one or more resource utilization policies.
14. The method of claim 8 , further comprising structuring a virtualization host specific mapping within the workload, wherein the resource management policy engine references the virtualization host specific mapping to request tuning the physical hardware resources allocated to the workload from one or more management domains associated with the one or more virtualization hosts.
15. The method of claim 14 , wherein the one or more management domains associated with the one or more virtualization hosts tune the physical hardware resources that the one or more virtualization hosts allocate to the workload in response to the one or more virtualization hosts having available physical hardware resources sufficient to comply with the one or more limits or ranges declared in the one or more virtual machine configuration files.
16. A non-transitory computer-readable storage medium containing a self-provisioning workload deployed to one or more virtualization hosts, wherein the self-provisioning workload contained in the non-transitory computer-readable storage medium comprises:
resource utilization instrumentation structured to collect utilization metrics from one or more physical hardware resources that the one or more virtualization hosts allocate to the workload;
a resource management policy engine structured to communicate with the one or more virtualization hosts to request tuning the physical hardware resources allocated to the workload in response to the utilization metrics collected with the resource utilization instrumentation violating one or more resource utilization policies defined to manage the workload; and
a virtualization host specific mapping, wherein the resource management policy engine references the virtualization host specific mapping to request tuning the physical hardware resources allocated to the workload from one or more management domains associated with the one or more virtualization hosts, wherein a packaged software stack injected into the self-provisioning workload structures the resource utilization instrumentation at a first level within the self-provisioning workload, the resource management policy engine at a second level above the resource utilization instrumentation, and the virtualization host specific mapping at a third level above the resource management policy engine.