Predictive virtual server scheduling and optimization of dynamic consumable resources to achieve priority-based workload performance objectives

Patent metadata

Field	Value
Publication number	US-2018165111-A1
Application number	US-201715434091-A
Country	US
Kind code	A1
Filing date	Feb 16, 2017
Priority date	Dec 8, 2016
Publication date	Jun 14, 2018
Grant date	—

Abstract

Embodiments include a technique for predictive virtual server scheduling and optimization of dynamic consumable resources to achieve priority-based workload performance objectives is provided. Techniques include calculating a capacity requirement for each virtual server based on an administrator defined performance objective for each virtual server. Techniques also include calculating an available CPU capacity for each importance level based on the calculated capacity requirement for each virtual server, sending the calculated available CPU capacity to a scheduler; and selecting a target hypervisor based on the calculated capacity requirement of each virtual server and the calculated available CPU capacity at each importance level of each hypervisor.

First claim

What is claimed is:

1 . A computer-implemented method for predictive virtual server scheduling and optimization of dynamic consumable resources to achieve priority-based workload performance objectives, the method comprising:
calculating a capacity requirement for each virtual server based on an administrator defined performance objective for each virtual server; calculating an available CPU capacity for each importance level based on the calculated capacity requirement for each virtual server; sending the calculated available CPU capacity to a scheduler; and selecting a target hypervisor based on the calculated capacity requirement of each virtual server and the calculated available CPU capacity at each importance level of each hypervisor.

2 . The method of claim 1 , wherein the selecting the target hypervisor selects a hypervisor with a most available CPU capacity at a lowest importance level for allocating the virtual server.

3 . The method of claim 1 , wherein the available CPU capacity A j h of a particular hypervisor h at an importance level j is calculated as follows:
A j h =A j-1 h −Σ i u j,i g

where A 1-1 h is the available CPU capacity of the hypervisor h at an importance level j- 1 and u j,i g is the capacity requirement to meet performance objective g on a particular virtual server i at an importance level j.

4 . The method of claim 1 , further comprising:
determining one or more non-performing virtual servers and an associated importance level of the one or more non-performing virtual servers; if one or more non-performing virtual servers are determined, determining a total needed capacity required by the non-performing virtual servers at the importance level to meet the performance objective; comparing the total needed capacity with an available capacity of a lower importance level; redistributing the available capacity from the lower importance level to the importance level of the non-performing virtual servers; comparing the total needed capacity to an updated available capacity of the importance level of the one or more non-performing virtual servers; and determining the one or more non-performing virtual servers as candidates for migration.

5 . The method of claim 1 , further comprising receiving a CPU capacity request for a virtual server at an importance level;
determining to place the virtual server on a hypervisor; reserving resources based on the CPU capacity request; updating an available CPU capacity at the importance level for the hypervisor based on the CPU capacity request; responsive to actual CPU usage of the virtual in the hypervisor, subsequently updating the available CPU capacity based on the actual CPU usage.

6 . The method of claim 1 , further comprising:
responsive to migrating or deleting a virtual server from a hypervisor, releasing resources based on a CPU capacity request for the virtual server; updating the available CPU capacity at an importance level for the hypervisor based on a last reported use for the hypervisor; and periodically updating the available CPU capacity of the hypervisor based on the actual CPU.

Assignees

• IBM

Inventors

• Feng Yong
• Gunal Yuksel
• Hu Xiao Liang
• Yocom Peter B

Classifications

• G06F9/4831 Primary

  with variable priority · CPC title

• G06F9/5077 Primary

  Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455) · CPC title

• G06F9/505

  considering the load · CPC title

• G06F9/4887

  involving deadlines, e.g. rate based, periodic · CPC title

• G06F9/45558 Primary

  Hypervisor-specific management and integration aspects · CPC title