Emc IP Holding Company LLC

Dell Products Inc.

Roberto Nery Stelling Neto

Rômulo Teixeira De Abreu Pinho

Vinicius Michel Gottin

Hugo De Oliveira Barbalho

Tiago Salviano Calmon

Pablo Nascimento Da Silva

Paulo Abelha Ferreira

Alex Laier Bordignon

Daniel Sadoc Menasché

Rodrigo Rios Almeida De Souza

Vitor Silva Sousa

Jonas Furtado Dias

Vinícius Michel Gottin

Ravi Shukla

Adriana Bechara Prado

Eduardo Vera Sousa

Jeffrey Scott Vah

Romulo D Pinho

Aaron Sanchez

Vítor Silva Sousa

One example method includes accessing input data elements from logs that identify user problems with computing system components, the data elements each associated with a respective original class label that identifies a class of computing system components to which the data element relates, the respective original class labels forming a group of class labels, and a first of the original class labels is overrepresented in the group, and reducing overrepresentation of the first original class label in the group by creating an arbitrary aggregation of some of the class labels that includes the first original class label. The method includes creating, based on a hierarchical modeling structure, prepared data in which an original class label is replaced by the aggregation. Next a hierarchical model and benchmark model are trained, and each model generates respective predictions for comparison. An inferencing process is performed to determine which predicted label will be used.

Hierarchical modeling approach for digital repair parts prediction

Method for model updating in a federated learning environment, including distributing a current model to client nodes; receiving a first set of gradient sign vectors, wherein each gradient sign vector of the first set of gradient sign vectors is received from one client node; generating a first updated model based on the first set of gradient sign vectors; distributing the first updated model to the plurality of client nodes; storing a first shape parameter and a second shape parameter; receiving, in response to distributing the first updated model, a second set of gradient sign vectors, wherein each gradient sign vector of the second set of gradient sign vectors is received from one client node; generating a second updated model based on the second set of gradient sign vectors, the first shape parameter, and the second shape parameter; and distributing the second updated model to the plurality of client nodes.

Fast converging gradient compressor for federated learning

One example method includes dynamically monitoring a stream of image portions that have been classified by a segmentation model of a video surveillance system, evaluating the image portions, based on the evaluating, determining that an attack on the video surveillance system is occurring, or has occurred, and implementing, or causing the implementation of, a remedial action with regard to the attack. The image portions may be image portions that have been classified by a segmentation model.

Delayed inference attack detection for image segmentation-based video surveillance applications

A framework for rapidly prototyping federated learning algorithms. Specifically, the disclosed framework proposes a method and system for evaluating different hypotheses for configuring learning model, which may be optimized through federated learning. Through the disclosed framework, these hypotheses may be tested for scalability, hardware and network resource performance, as well as for new learning state compression and/or aggregation technique effectiveness. Further, these hypotheses may be tested through federated learning simulations, which avoid costs associated with deploying these hypotheses to be tested across production systems.

Framework for rapidly prototyping federated learning algorithms

Methods for compression switching that includes distributing a model to client nodes, which use the model to generate a gradient vector (GV) based on a client node data set. The method includes receiving a model update that includes a gradient sign vector (GSV) based on the gradient vector; generating an updated model using the GSV; and distributing the updated model to the client nodes. The client node uses the updated model to generate a second GV based on a second client node data set. The method also includes a determination that a compression switch condition exists; based on the determination, transmitting an instruction to the client node to perform a compression switch; receiving, in response to the instruction, another model update including a subset GSV based on the second gradient vector; generating a second updated model using the subset GSV; and distributing the second updated model to the client nodes.

Compression switching for federated learning

Facilitating detection of anomalies of a target entity is provided herein. A system can comprise a processor and a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations. The operations can comprise training a model on a first set of variables that are constrained by a second set of variables. The second set of variables can characterize elements of a defined entity. The first set of variables can define a normality of the defined entity. The operations also can comprise employing the model to identify expected parameters and unexpected parameters associated with the defined entity to at least a defined level of confidence.

Facilitating detection of anomalies in data center telemetry

Facilitating temporal data management for anomalous state detection in data centers is provided herein. A system can comprise a processor and a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations. The operations can comprise performing a process of extraction, transformation, and loading of data from log files into a telemetry data store. The data can be loaded into the telemetry data store as telemetry data. The operations also can comprise dividing the telemetry data into first telemetry data and second telemetry data. The first telemetry data can comprise telemetry data that does not satisfy a defined quality level. The second telemetry data can comprise telemetry data that satisfies the defined quality level. Further, the operations can comprise removing the first telemetry data from the telemetry data store and outputting the second telemetry data based on a request for the second telemetry data.

Facilitating temporal data management for anomalous state detection in data centers

An information handling system for managing a storage system includes storage for storing profile-based cache policy performance prediction models. The information handling system also includes a storage manager that obtains an input-output profile for a workload hosted by the information handling system during a first period of time; obtains performance metrics for cache policies for the storage system using: the input-output profile, and the profile-based cache policy performance prediction models; obtains a ranking of the cache policies based on the performance metrics; selects a cache policy of the cache policies based on the rankings; and updates operation of a cache of the storage system based on the selected cache policy for a second period of time.

System and method for cache management

A cache management system includes a sequentiality determination process configured to determine sequentiality profiles of a workload of IO traces as the workload dynamically changes over time. A learning process is trained to learn a correlation between workload sequentiality and cache pollution, and the trained learning process is used to predict cache pollution before the cache starts to experience symptoms of excessive pollution. The predicted pollution value is used by a cache policy adjustment process to change the prefetch policy applied to the cache, to proactively control the manner in which prefetching is used to write data to the cache. Selection of the cache policy is implemented on a per-LUN basis, so that cache performance for each LUN is individually managed by the cache management system.

Method and apparatus for adjusting cache prefetch policies based on predicted cache pollution from dynamically evolving workloads

Techniques are provided for characterizing and quantifying a sequentiality of workloads using sequentiality profiles and signatures. One exemplary method comprises obtaining telemetry data for an input/output workload; evaluating a distribution over time of sequence lengths for input/output requests in the telemetry data by the input/output workload; and generating a sequentiality profile for the input/output workload to characterize the input/output workload based at least in part on the distribution over time of the sequence lengths. Multiple sequentiality profiles for one or more input/output workloads may be clustered into a plurality of clusters. A sequentiality signature may be generated to represent one or more sequentiality profiles within a given cluster. A performance of data movement policies may be evaluated with respect to the sequentiality signature of the given cluster.

Sequentiality characterization of input/output workloads

A pre-fetching technique determines what data, if any, to pre-fetch on a per-logical storage unit basis. For a given logical storage unit, what, if any, data to prefetch is based at least in part on a collective sequential proximity of the most recently requested pages of the logical storage unit. Determining what, if any, data to pre-fetch for a logical storage unit may include determining a value for a proximity metric indicative of the collective sequential proximity of the most recently requested pages, comparing the value to a predetermined proximity threshold value, and determining whether to pre-fetch one or more pages of the logical storage unit based on the result of the comparison. A data structure may be maintained that includes most recently requested pages for one or more logical storage units. This data structure may be a table.

Growing community of inventors

Rio de Janeiro, Brazil

Roberto Nery Stelling Neto