Microsoft Technology Licensing, LLC

Ashish Munjal

Lee William Progl

Christopher G Kaler

Prasad Durga Ayyalasomayajula

Uuganjargal Khanna

Laura Marie Caulfield

Systems and methods for searching for hash strings stored in an indexed array are described. A method includes receiving a hash string. The method includes determining a first set of index values for the indexed array that correspond to a first stored value matching a first portion of the hash string and determining a second set of index values for the indexed array that correspond to a second stored value matching a second portion of the hash string if a match between the first stored value and the first portion of the hash string is found. The method includes upon finding a match for the first stored value and the second stored value, comparing the hash string to each of the set of hash strings in the indexed array having an index value common to both the first set of index values and the second set of index values.

Searching for a hash string stored in an indexed array

Systems and methods for creating and deploying packages to devices in a fleet based on operations derived from a machine learning model or other automated feedback models are provided. As an example, a method for creating a package, including a payload, for deployment to a set of devices is provided. The method includes receiving a payload, where the payload has an associated set of payload parameters concerning a deployment of the payload to the set of devices. The method further includes using a processor, automatically creating the package for the deployment to the set of devices, where the package comprises instructions for deploying the payload to the set of devices, and where the instructions specify at least one of a plurality of operations derived from a machine learning model based at least on a subset of the associated set of payload parameters.

Creating and deploying packages to devices in a fleet based on operations derived from a machine learning model

Systems and methods for deploying packages to devices in a fleet in stages are provided. A method includes first scanning hardware configured to handle functions unrelated to the customer workloads in a first stage to determine whether a selected diversity target for deploying the package is met. The method further includes, if the selected diversity target for deploying the package is not met, then scanning hardware configured to handle at least a subset of the customer workloads in a second stage to determine whether the selected diversity target for deploying the package is met. The method further includes, if the selected diversity target for deploying the package is met after the scanning, then using the processor, deploying the package to the first subset of the set of devices in the first stage and to a second subset of the set of devices in the second stage.

Deploying packages to devices in a fleet in stages

A fully out-of-band, automated process provides centralized, concurrent in-system cluster-level/hyper-scale CPU thermal excursion signature detection based on system event log (SEL) content. A chassis manager within a hyper-scale cluster operates as the initiator of a cluster-level SEL collection and analysis operation. The master chassis manager initiates a collect and analyze operation, which gets relayed to every chassis manager within the cluster. Every recipient chassis manager then further propagates the collect and analyze operation request to each of the server blades hosted within the chassis. A Baseboard Management Controller (BMC) on each server hosted within the chassis receives a cluster SEL read request from the chassis manager through out-of-band communication interfaces. The BMC forwards SEL data for the server back to the chassis manager, which analyzes the data received from all the servers on the chassis and then sends the results to next manager in the hierarchy.

Thermal excursion detection in datacenter components

Techniques of implementing out-of-band data erasure verification in computing systems are disclosed herein. In one embodiment, a method includes receiving a verification instruction from a system administrator. In response to and based on the received verification instruction, the method includes selecting a set of persistent storage devices to which data erasure verification is to be performed. The method also includes relaying the verification instruction to additional computing devices in additional enclosures, thereby causing data erasure verification on one or more additional persistent storage devices in the additional enclosures be performed generally in parallel to performing data erasure verification on the subset of persistent storage devices in one of the enclosures.

Secure data erasure verification in hyperscale computing systems

Illustrated herein is a substantially circular data center including a plurality of machines. The machines are physically grouped into data classification groups. The data classification groups are organized in a radially expanding and circumferential fashion.

Growing community of inventors

Bellevue, WA, United States of America

Ashish Munjal