17806191. SYSTEM AND METHOD OF DYNAMICALLY ADJUSTING VIRTUAL MACHINES FOR A WORKLOAD simplified abstract (Microsoft Technology Licensing, LLC)

SYSTEM AND METHOD OF DYNAMICALLY ADJUSTING VIRTUAL MACHINES FOR A WORKLOAD

Organization Name

Microsoft Technology Licensing, LLC

Inventor(s)

Soumya Ram of Cambridge MA (US)

Preston Tapley Stephenson of Boston MA (US)

Alexander David Fischer of Salt Lake City UT (US)

Mahmoud Sayed of Hyattsville MD (US)

Robert Edward Minneker of Seattle WA (US)

Eli Cortex Custodio Vilarinho of Woodinville WA (US)

Felipe Vieira Frujeri of Kirkland WA (US)

Inigo Goiri Presa of Redmond WA (US)

Sidhanth M. Panjwani of New York NY (US)

Yandan Wang of Kirkland WA (US)

Camille Jean Couturier of Nantes (FR)

Jue Zhang of Beijing (CN)

Fangkai Yang of Beijing (CN)

Si Qin of Beijing (CN)

Qingwei Lin of Beijing (CN)

Chetan Bansal of Seattle WA (US)

Bowen Pang of Beijing (CN)

Vivek Gupta of Groton MA (US)

SYSTEM AND METHOD OF DYNAMICALLY ADJUSTING VIRTUAL MACHINES FOR A WORKLOAD - A simplified explanation of the abstract

This abstract first appeared for US patent application 17806191 titled 'SYSTEM AND METHOD OF DYNAMICALLY ADJUSTING VIRTUAL MACHINES FOR A WORKLOAD

Simplified Explanation

The abstract describes a method for dynamically adjusting the number of virtual machines for a workload based on the likelihood of eviction during different stages. The method involves receiving probability indicators for each stage, predicting the target number of virtual machines for the subsequent stage, and configuring the number of virtual machines based on the target number.

• The method is used to dynamically adjust the number of virtual machines for a workload.
• Probability indicators are received for each stage to determine the likelihood of eviction.
• The target number of virtual machines for the subsequent stage is predicted based on the probability indicator, target capacity, and current price.
• The number of virtual machines for the workload is configured during the current stage based on the target number for the subsequent stage.

Potential Applications

• Cloud computing: This method can be applied in cloud computing environments to optimize the allocation of virtual machines for workloads.
• Resource management: It can be used in resource management systems to efficiently allocate virtual machines based on workload demands.
• Cost optimization: By dynamically adjusting the number of virtual machines, this method can help reduce costs associated with maintaining unnecessary virtual machines.

Problems Solved

• Inefficient resource allocation: This method addresses the problem of inefficiently allocating virtual machines for workloads by dynamically adjusting their numbers based on eviction likelihood.
• Cost inefficiency: By considering the current price for maintaining a virtual machine, this method helps optimize costs by only configuring the necessary number of virtual machines.

Benefits

• Improved resource utilization: By dynamically adjusting the number of virtual machines, this method ensures optimal utilization of resources for workloads.
• Cost savings: The method helps reduce costs by only configuring the necessary number of virtual machines based on workload demands and eviction likelihood.
• Enhanced workload performance: By optimizing the allocation of virtual machines, this method can improve the performance and responsiveness of workloads.

Original Abstract Submitted

A method for dynamically adjusting a number of virtual machines for a workload, includes: receiving a probability indicator for each of a plurality of N sequential stages, where N is a natural number greater than 1, of a likelihood that a virtual machine assigned to a workload will be evicted during the N sequential stages; predicting a target number of virtual machines to configure in a current stage for a subsequent stage from among the plurality of N sequential stages based on the probability indicator, a target capacity for the workload, and a current price for maintaining a virtual machine; and configuring a number of virtual machines for the workload during the current stage based on the target number to be loaded for the workload for the subsequent stage.