# US 2021/0240536 A1 · Allocation and management of computing platform resources

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0240536 A1
- Document ID: US20210240536A1
- Status: Published application
- Kind: A1
- Application number: 16/865,848
- Filed: 2020-05-04
- Published: 2021-08-05
- Earliest claimed: 2020-01-31
- Assignees: salesforce.com, inc.
- Applicants: salesforce.com, inc.
- Inventors: Jeremiah David Brazeau; Sumanth Kondakindi; Dmitriy Ryabkov; Adam Elliott; Stefan Piesche; Metin Kilic; Yan Huang; Christopher David Boran
- Web page: https://patentgrep.com/patent/US20210240536A1

## CPC

- G06F 9/5027: {the resource being a machine, e.g. CPUs, Servers, Terminals}
- G06F 11/3006: {where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems (multiprogramming arrangements G06F9/46; allocation of resources G06F9/50)}
- G06F 11/302: {where the computing system component is a software system}
- G06F 11/3409: {for performance assessment}
- G06F 11/3433: {for load management (allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029)}
- G06F 2201/81: Threshold
- G06F 2209/5018: Thread allocation
- G06F 2209/504: Resource capping
- G06F 2209/506: Constraint
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Systems and techniques are provided for monitoring and managing the performance of services accessed by sites on a computing platform. When a performance issue is identified, a service is monitored to determine if calls to the service exceed a threshold completion time. If so, a resource available to call the service is adaptively throttled by the platform.

## Claims

### Claim 1

A computer-implement method comprising:
determining, by a computing platform that provides computing resources to a plurality of sites that are accessed by a plurality of users and that manages access by the sites to a first service, that the first service is not meeting at least one performance metric;
in response to determining that the first service is not meeting the at least one performance metric, determining whether the first service is thread-limited by the platform;
where the first service is not thread-limited by the platform:
setting a limit on the number of threads available for calls to the first service by sites on the platform;
monitoring performance of the service after setting the limit on the number of threads available for calls to the first service by sites on the platform and, based upon the performance of the service after setting the limit on the number of threads, changing the limit on the number of threads available for calls to the first service by sites on the platform; and
where the first service is thread-limited by the platform:
changing the limit on the number of threads available for calls to the first service.

### Claim 2 (depends on 1)

The method of claim 1, wherein the performance metric is based on a typical response time of calls made to the service.

### Claim 3 (depends on 2)

The method of claim 2, wherein the performance metric is a weighted average of response times of calls made to the service.

### Claim 4 (depends on 1)

The method of claim 1, wherein the service is a resource that is external to the computing platform.

### Claim 5 (depends on 1)

The method of claim 1, wherein the service is a resource that is provided by the computing platform, which is separate and distinct from the plurality of sites.

### Claim 6 (depends on 1)

The method of claim 1, wherein the first service is an external resource that is identified by the computing platform based on a plurality of direct calls made to the external resource by the plurality of sites on the platform.

### Claim 7 (depends on 6)

The method of claim 6, further comprising identifying a sub-service within the first service that is provided by a separate portion of the external resource based upon common portions of a uniform resource locator (URL) used to access the first service and the sub-service by the plurality of sites.

### Claim 8

A computing platform that provides computing resources to a plurality of sites that are accessed by a plurality of users and that manages access by the sites to a first service, the computing platform comprising:
a processor; and
a computer-readable memory storing instructions for execution by the processor which, when executed by the processor, cause the processor to:
determine that the first service is not meeting at least one performance metric;
in response to determining that the first service is not meeting the at least one performance metric, determine whether the first service is thread-limited by the platform;
where the first service is not thread-limited by the platform:
set a limit on the number of threads available for calls to the first service by sites on the platform;
monitor performance of the service after setting the limit on the number of threads available for calls to the first service by sites on the platform and, based upon the performance of the service after setting the limit on the number of threads, change the limit on the number of threads available for calls to the first service by sites on the platform; and
where the first service is thread-limited by the platform:
change the limit on the number of threads available for calls to the first service.

### Claim 9 (depends on 8)

The computing platform of claim 8, wherein the performance metric is based on a typical response time of calls made to the service.

### Claim 10 (depends on 9)

The computing platform of claim 9, wherein the performance metric is a weighted average of response times of calls made to the service.

### Claim 11 (depends on 9)

The computing platform of claim 9, wherein the service is a resource that is external to the computing platform.

### Claim 12 (depends on 9)

The computing platform of claim 9, wherein the service is a resource that is provided by the computing platform, which is separate and distinct from the plurality of sites.

### Claim 13 (depends on 8)

The computing platform of claim 8, wherein the first service is an external resource that is identified by the computing platform based on a plurality of direct calls made to the external resource by the plurality of sites on the platform.

### Claim 14 (depends on 13)

The computing platform of claim 13, further comprising identifying a sub-service within the first service that is provided by a separate portion of the external resource based upon common portions of a uniform resource locator (URL) used to access the first service and the sub-service by the plurality of sites.

### Claim 15

A method comprising:
calculating a plurality of performance levels for a first service of a plurality of services accessed by a plurality of sites on a computing platform that manages access to the plurality of services by the plurality of sites, each performance level indicating a response time below which the majority of calls to the service are completed;
determining a maximum service call completion threshold based upon the plurality of performance levels;
determining that a predefined percentage of calls to the first service within a predefined period of time exceed the maximum service call completion threshold and, in response, applying a throttle to a resource used to call the service by sites on the platform.

### Claim 16 (depends on 15)

The method of claim 15, wherein the performance level is a response time below which 95% of calls to the first service within a predefined period of time are completed.

### Claim 17 (depends on 15)

The method of claim 15, wherein the resource used to call the service by sites on the platform is a number of threads available for calling the service by sites on the platform.

### Claim 18 (depends on 17)

The method of claim 17, wherein applying the throttle comprises setting a maximum number of threads available for calling the service by sites on the platform.
