Training and using a memory failure prediction model

US12379983B2 · kind B2 · utility

Assignee

• MICROSOFT TECHNOLOGY LICENSING, LLC · US

Inventors

• Jasmine Grace Schlichting · Seattle, US
• Bhuvan Malladihalli Shashidhara · Bellevue, US
• Ramakoti Reddy Bhimanadhuni · Suwanee, US
• Emily N. Wilson · Seattle, US
• Farah Farzana · Redmond, US
• Michael Wayne Stephenson · Woodinville, US
• Pallavi Baral · Redmond, US
• Josh Charles Moore · Lynnwood, US
• Christina Margaret Tobias · Seattle, US
• John A. Strange · Rye, US
• Peter Hanpeng Jiang · Kirkland, US
• Sebastien Nathan R Levy · Seattle, US
• Brett K. Dodds · Boise, US
• Arhatha Bramhanand · Redmond, US
• Juan Arturo Herrera Ortiz · Seattle, US
• Ahu Oral · Seattle, US
• Charlotte Gauchet · Suresnes, FR
• Daniel Sebastian BERGER · Philadelphia, US

Key dates

• Filing date: Nov 19, 2021
• Grant date: Aug 5, 2025
• Expiry date: Jun 6, 2044

Classification

• Technology area (CPC G): Physics
• CPC primary: G06F11/3058
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

The disclosure herein describes training and using an uncorrectable error (UE) state prediction model based on telemetry error data. Sets of UE state labels and non-UE state labels are generated from a first set of collected telemetry data, wherein the UE state labels each reference a UE and telemetry data of an interval prior to the referenced UE. Statistical features are extracted from telemetry data of the sets of UE state labels and non-UE state labels, and the extracted statistical features are used to train a UE state prediction model. A second set of collected telemetry data is obtained, and a UE event is predicted based on the second set of collected telemetry data using the trained UE state prediction model. A preventative operation is performed on a memory page of the system based on the predicted UE event, whereby the predicted UE event is prevented from occurring.