IP Library Granted Patent US 12,032,883
Granted Patent B1
US 12,032,883 · App. 18/333,654 · Granted Jul 9, 2024

Optimization of configurable distributed computing systems

Inventors: Vasileios Koutsomanis (Moscow, RU); Igor Marnat (Moscow, RU); Nikolay Dobrovolskiy (Moscow, RU)
Assignee: Parallels International GmbH
G06F30/20G06F9/452G06F9/5027G06F9/5077
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,032,883
App. No.
18/333,654
Filed
Jun 13, 2023
Granted
Jul 9, 2024
Kind
B1
Examiner
HO, ANDY
Art Unit
2194
USPC
703/21
Abstract

The subject matter of this specification can be implemented in, among other things, a method that includes accessing a plurality of target tasks for a computing system, the computing system comprising a plurality of resources, wherein the plurality of resources comprises a first server and a second server, accessing a plurality of configurations of the computing system, wherein each of the plurality of configurations identifies one or more resources of the plurality of resources to perform the respective target task of the plurality of target tasks, and performing, for each of the plurality of configurations, a simulation to determine a plurality of performance metrics, wherein each of the plurality of performance metrics predicts performance of at least one of the plurality of resources executing the plurality of target tasks on the computing system.

Claims (26)

1. A method comprising:

establishing, by a processing device, a plurality of configurations of a computing system where each configuration of the plurality of configurations identifies a subset of a plurality of resources of the computing system;

establishing, by the processing device, a plurality of task sets where each task set of the plurality of task sets defines a subset of a plurality of target tasks to be executed by the computing system; and

performing, for each configuration of the plurality of configurations and each task set of the plurality of task sets, a simulation to determine a plurality of performance metrics, wherein each of the plurality of performance metrics predicts performance of at least one resource of the subset of the plurality of resources for that configuration of the plurality of configurations executing that task set of the plurality of task sets.

2. The method according to claim 1 , further comprising

determining, with a configuration analysis module, for each configuration of the plurality of configurations a task set of the plurality of task sets to be associated with that configuration of the plurality of configurations of the computing system in dependence upon overall performance metrics for the plurality task sets simulated with that configuration of the plurality of configurations; wherein

each overall performance metric for a task set of the plurality of task sets for a configuration of the plurality of configurations is established in dependence upon the plurality of performance metrics determined by the simulation of that task set of the plurality of task sets for that configuration of the plurality of configurations.

3. A method comprising:

executing a configuration analysis module upon one or more processors forming part of a distributed computing system comprising a plurality of resources;

retrieving from a memory accessible by the configuration analysis module a plurality of target tasks to be executed by the distributed computing system;

retrieving from another memory accessible by the configuration analysis module a plurality of configurations of the distributed computing system,

performing, with the configuration analysis module, a plurality of simulations to determine a plurality of performance metrics where:

each simulation of the plurality of simulations identifies a subset of the plurality of target tasks and a subset of the plurality of resources to be applied to the subset of the plurality of target tasks; and

each performance metric of the plurality of performance metrics predicts performance of at least one resource of subset of the plurality of resources to be employed in executing the subset of the plurality of target tasks upon the distributed computing system;

generating for each simulation of the plurality of simulations an overall performance metric in dependence upon the plurality of performance metrics established for that simulation of the plurality of simulations;

determining, with the configuration analysis module, a plurality of system configurations in dependence upon the overall performance metrics for the plurality simulations where each system configuration of the plurality of system configurations defines a subset of the plurality of target tasks and that subset of the plurality of resources which upon being applied to the subset of the plurality of target tasks yield the highest overall performance metric.

4. The method according to claim 3 , further comprising

establishing a required set of tasks to be performed in association with a remote session upon the distributed computer system;

identifying, with the configuration analysis module, the system configuration of the plurality of system configurations where the subset of the plurality of target tasks simulated for that system configuration of the plurality of system configurations matches the required set of tasks;

configuring, with the configuration analysis module, a portion of the distributed computing system in dependence upon the identified configuration of the plurality of configurations.

5. The method according to claim 3 , wherein

execution of the configuration analysis module to perform the simulations for the plurality of configurations is initiated upon the installation of a software application to the distributed computing system.

6. The method according to claim 3 , wherein

execution of the configuration analysis module to perform the simulations for the plurality of configurations is initiated by the distributed computing system prior to the distributed computing system providing one or more remote sessions to one or more users of the distributed computing system.

7. The method according to claim 4 , wherein

execution of the configuration analysis module to perform the simulations for the plurality of configurations is initiated upon the initiation of the remote session.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 20, 2023
From: KOUTSOMANIS, VASILEIOS; MARNAT, IGOR; DOBROVOLSKIY, NIKOLAY
To: PARALLELS INTERNATIONAL GMBH
Reel/Frame 063995/0717 →
Continuity (2)
Continuation 17739864 · May 9, 2022
Continuation 16837305 · Apr 1, 2020
Cited By (1)
US 12,367,125