US 2021/0240534 A1 · ADAPTIVE MEMORY OPTIMIZATION FOR THIN CLIENTS

Abstract

A memory optimizer service can be used to provide adaptive memory optimization. The memory optimizer service can be executed on a thin client or other user device to detect when low memory conditions exist. When it detects a low memory condition, the memory optimizer service can obtain process memory information for processes executing on the thin client and use the process memory information to calculate a trim amount to apply to the working set of processes with the highest amount of private bytes in their working sets. These trim amounts can be proportional to the amounts of private bytes and in sum can equal a total desired reduction in memory consumption.

Claims20

Dependent claims

A method, implemented by a memory optimizer service that executes on a user device, for performing adaptive memory optimization, the method comprising:

detecting a low memory condition;
in response to detecting the low memory condition, obtaining process memory information that includes, for each of a plurality of processes, a size of a working set for the process and an amount of private bytes in the working set for the process;
identifying a subset of the plurality of processes;
for each process in the subset, calculating a trim amount to apply to the working set of the process, the trim amount being based on the amount of private bytes in the working set for the process;
and
for each of at least some of the processes in the subset, causing the working set for the process to be reduced by the corresponding trim amount.

3 more elements

One or more computer storage media storing computer executable instructions which when executed on a user device implement a memory optimizer service that is configured to implement a method for performing adaptive memory optimization, the method comprising:

detecting a low memory condition;
in response to detecting the low memory condition, obtaining process memory information that includes, for each of a first plurality of processes, a size of a working set for the process and an amount of private bytes in the working set for the process;
identifying processes that have higher amounts of private bytes than other processes in the first plurality of processes;
identifying a total desired reduction in memory consumption;
calculating trim amounts to apply to the working sets of the identified processes to achieve the total desired reduction in memory consumption, each trim amount being proportional to the amount of private bytes in the working set for the corresponding process;
and
causing the working set of each identified process to be reduced by the corresponding trim amount.

4 more elements

Dependencies are inferred from the claim text.

Adaptive memory optimization for thin clients

Life of this patent

Adaptive memory optimization for thin clients

Life of this patent

Abstract

Claims20