# US 2024/0004630 A1 · Applying service level objectives to release pipelines

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

- Number: US 2024/0004630 A1
- Document ID: US20240004630A1
- Status: Published application
- Kind: A1
- Application number: 17/855,123
- Filed: 2022-06-30
- Published: 2024-01-04
- Earliest claimed: 2022-06-30
- Assignees: Microsoft Technology Licensing, LLC
- Applicants: Microsoft Technology Licensing, LLC
- Inventors: Jeremy Michael CHANDRA; Havva Gulay GURBUZ; Marc VAN EIJK; Rajive KUMAR; Jay Christopher SVOBODA; Kiran Belthur PARTHASARATHY
- Web page: https://patentgrep.com/patent/US20240004630A1

## CPC

- G06F 8/65: Updates (security arrangements therefor G06F21/57)
- G06F 8/71: Version control (security arrangements therefor G06F21/57); Configuration management
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

The present disclosure relates to methods and systems for monitoring release pipelines. The methods and systems calculate health metrics of release pipelines using telemetry data from the release pipelines. The methods and systems determine whether the release pipelines are healthy based on the health metrics. The methods and systems automatically trigger one or more actions in response to determining that the release pipelines are unhealthy and continue to monitor the release pipelines in response to determining that the release pipelines are healthy.

## Claims

### Claim 1

A method, comprising:
calculating health metrics of a release pipeline comprising a stage using telemetry data from the release pipeline, wherein the telemetry data is obtained during execution of a release by the release pipeline;
determining that the release pipeline is unhealthy based on the health metrics; and
triggering an action in response to determining that the release pipeline is unhealthy.

### Claim 2 (depends on 1)

The method of claim 1, wherein the action is sending an alert with a notification that the release pipeline is unhealthy.

### Claim 3 (depends on 1)

The method of claim 1, wherein the action is preventing additional releases from being executed by the release pipeline.

### Claim 4 (depends on 1)

The method of claim 1, wherein determining that the release pipeline is unhealthy includes comparing the health metrics to a respective threshold.

### Claim 5 (depends on 4)

The method of claim 4, wherein if the health metrics is below the respective threshold, determining that the release pipeline is unhealthy.

### Claim 6 (depends on 4)

The method of claim 4, wherein if the health metrics is equal to the respective threshold or exceeds the respective threshold, determining that the release pipeline is unhealthy.

### Claim 7 (depends on 1)

The method of claim 1, wherein the health metrics include a latency of a release that provides a duration in time for a release instance of the release to go from a test stage to a last stage of the release pipeline.

### Claim 8 (depends on 1)

The method of claim 1, wherein the health metrics include a success rate of a release in the release pipeline based on a total number of successful stages in the release relative to a total number of stages in the release.

### Claim 9 (depends on 1)

The method of claim 1, wherein the health metrics include a throughput of the release pipeline based on a number of release instances of a release that complete a last stage of the release pipeline.

### Claim 10 (depends on 1)

The method of claim 1, wherein the health metrics include a drift score of the release pipeline that identifies a recency of a release instance.

### Claim 11

A method, comprising:
executing a plurality of releases on a plurality of release pipelines;
calculating health metrics for each release pipeline using telemetry data from each release pipeline; and
performing an action in response to determining that a release pipeline of the plurality of release pipelines is unhealthy based on the health metrics.

### Claim 12 (depends on 11)

The method of claim 11, wherein each release pipeline of the plurality of release pipelines corresponds to a separate release of the plurality of releases.

### Claim 13 (depends on 11)

The method of claim 11, wherein each release pipeline includes a plurality of stages for each release of the plurality of releases.

### Claim 14 (depends on 11)

The method of claim 11, wherein the telemetry data is obtained during an execution of each release on each release pipeline.

### Claim 15 (depends on 11)

The method of claim 11, wherein the health metrics include a latency of each release, and wherein determining that the release pipeline is unhealthy further comprises:
comparing the latency of each release to a threshold for latency; and
if the latency of the release pipeline exceeds the threshold for latency, determining that the release pipeline is unhealthy.

### Claim 16 (depends on 11)

The method of claim 11, wherein the health metrics include a success rate of each release, and wherein determining that the release pipeline is unhealthy further comprises:
comparing the success rate to a threshold for the success rate; and
if the success rate of the release pipeline is below the threshold for the success rate, determining that the release pipeline is unhealthy.

### Claim 17 (depends on 11)

The method of claim 11, wherein the health metrics include a throughput of each release pipeline, and wherein determining that the release pipeline is unhealthy further comprises:
comparing the throughput to a threshold for the throughput; and
if the throughput of the release pipeline exceeds the threshold for the throughput, determining that the release pipeline is unhealthy.

### Claim 18 (depends on 11)

The method of claim 11, wherein the health metrics include a drift score of each release pipeline and determining that the release pipeline is unhealthy further comprises:
comparing the drift score to a threshold for the drift score; and
if the drift score of the release pipeline exceeds the threshold for the drift score, determining that the release pipeline is unhealthy.

### Claim 19 (depends on 11)

The method of claim 11, wherein the action is sending an alert with a notification identifying which release pipeline of the plurality of release pipelines is unhealthy.

### Claim 20 (depends on 11)

The method of claim 11, wherein the action is preventing additional releases from occurring on the release pipeline.
