Emc IP Holding Company LLC

Emc Corporation

Dell Products Inc.

Other

John T Cardente

John S Harwood

Gaurav Chawla

Junping Zhao

Robert Anthony Lincourt, Jr

Layne Lin Peng

Yifan Sun

Michael Kowalchik

Kenneth Durazzo

Uday K Gupta

Yu Cao

Sanping Li

David Stephen Reiner

Frank Charles Paterra

John Forecast

Stephen A Fridella

Zachary Merlynn Loafman

Deepak Sunny Veliath

Kun Wang

Stuart Allen Berke

William Price Dawkins

Tiago Salviano Calmon

Kevin Xu

Vinícius Michel Gottin

Xiaoyan Guo

Douglas L Farley

Zhe Dong

Christopher H Stacey

Simon Tao

Accela Yilong Zhao

Michael B Evans

Tommi Salli

Qi Bao

Kirk Frey

James Theodore Compton

Jack Harwood

Christopher H E Stacey

Anna Groves

Chris Stacey, Jr

A data transformation/processing system includes a data transformation/processing management device coupled to a client device, processing systems, memory systems, and storage systems. The data transformation/processing management device receives a request from the client device to process first data to generate second data, determines a first subset of the processing systems for performing processing operations to process the first data, and identifies a data path for the processing operations that includes storage locations provided by subsets of the memory systems and the storage systems. The data transformation/processing management device then determines a second subset of the processing systems in the data path for performing transformation operations to transform the first data for processing and storage in the storage locations, and configures the first subset of the processing systems to perform the processing operations and the second subset of the processing systems to perform the transformation operations.

Data processing path data transformation system

A resource-capability-and-connectivity-based workload performance improvement system includes a resource management device coupled to processing systems and memory systems. The resource management device receives a request to perform a first workload, identifies a DAG that includes functions for performing the first workload, and uses first parameters in the DAG to configure the processing systems and the memory subsystems to perform the functions. Based on performance of the functions, the resource management device determines function modification(s) for at least one of the functions and, based on the function modification(s), modifies the first parameters included in the DAG to provide modified parameters. When the resource management device receives a request to perform a second workload, it identifies the DAG that includes the functions for performing the second workload, and uses the modified parameters in the DAG to configure the processing systems and the memory subsystems to perform the functions.

Resource-capability-and-connectivity-based workload performance improvement system

A method for generating a domain topology for a system, the method that includes accessing a functional model database, identifying, in a computing resource database, a configuration of components matching a functional model in the functional model database, generating a domain for the configuration of components, in a domain topology, and generating a graphical hierarchy of the domain topology and providing it to a user.

Systems and methods for identifying and organizing device capabilities into graphical hierarchies

A distributed function data transformation system includes a distributed function packet provisioning device coupled to compute systems. The distributed function packet provisioning device receives a request to perform a data transformation including data transformation operations, and determines a respective function for performing each data transformation operation. The distributed function packet provisioning device generates a distributed function packet including a function list identifying the respective functions, a data identifier identifying data upon which the respective functions should be performed, and a function performance identifier configured to identify one of the respective functions to perform on the data. The distributed function packet provisioning device then transmits the distributed function packet to a first compute system to cause it to perform a first function identified in the function list and transmit the distributed function packet to a second compute system to cause it to perform a second function identified in the function list.

Distributed function data transformation system

Methods and apparatus are provided for real-time anomaly detection over sets of time-series data. One method comprises: obtaining a state-space representation of a plurality of states and transitions between said states based on sets of historical time-series data; obtaining an anomaly detection model trained using a supervised learning technique, wherein the anomaly detection model associates sequences of states in the state-space representation with annotated anomalies in the sets of historical time-series data and assigns a probability to said sequences of states; and, for incoming real-time time-series data, determining a likelihood of a current state belonging to a plurality of possible states in the state-space representation; and determining a probability of incurring said annotated anomalies based on a plurality of likely current state sequences that satisfy a predefined likelihood criteria. Anomalous behavior is optionally distinguished from previously unknown behavior based on a predefined likelihood threshold.

Methods and apparatus for real-time anomaly detection over sets of time-series data

Techniques described herein relate to a method for managing training data. The method includes monitoring, by a training data stream manager (TDSM), a cache comprising a plurality of training data examples associated with streams of mini-batch sequences scheduled to be transmitted to a machine learning training environment; making a first determination that a cache eviction is required; in response to the first determination: selecting a training data example of the plurality of training data examples; making a second determination that the training data example is eligible for cache eviction; in response to the second determination: evicting the training data example from the cache; and updating a training data example database entry to indicate that the training data example is evicted from the cache.

Method and system for evicting and reloading a cache for machine learning training data streams

A data storage transformation system includes a data storage management device that is coupled to a data provisioning device and a storage system. The data storage management device receives first format data that includes a first data format from the data provisioning device, and predicts at least one processing operation that will be performed on the first format data. The data storage management device then determines a second data format for performing the at least one processing operation, and transforms the first format data to second format data that includes the second data format. The data storage management device then transmits the second format data for storage in the storage system.

Data storage transformation system

A tiered memory fabric workload performance optimization system includes a workload management device coupled to a processing fabric and a memory fabric. The workload management system receives a workload request to perform a workload including sub-workloads, and identifies a respective processing system in the processing fabric for performing each of the sub-workloads. The workload management device then determines, for use by each respective processing system identified for performing the sub-workloads, a respective memory system in the memory fabric to provide memory systems in different memory tiers in the memory fabric that optimize characteristic(s) of a workload performance pipeline provided by the respective processing systems identified for performing the sub-workloads. The workload management device then configures each respective processing system identified for performing each of the sub-workloads, and the respective first memory system determined for that respective processing system, to perform the sub-workload that respective processing system was identified to perform.

Tiered memory fabric workload performance optimization system

An apparatus in one embodiment comprises a plurality of container host devices of at least one processing platform. The container host devices implement a plurality of containers for executing applications on behalf of one or more tenants of cloud infrastructure. One or more of the container host devices are each configured to compute distance measures between respective pairs of the containers and to assign the containers to container clusters based at least in part on the distance measures. The distance measures may be computed as respective content-based distance measures between hash identifiers of respective layers of layer structures of the corresponding containers. The apparatus may further comprise an interface configured to present a visualization of the container clusters. User feedback received via the interface is utilized to alter at least one parameter of the computation of distance measures and the assignment of clusters to container clusters.

Container distance measurement and clustering

A time-series data cache is operatively coupled between a time-series analytics application program and a time-series data store, and configured to temporarily store portions of the time-series data. The time-series data store is configured to persistently store time-series data. The time-series data cache is further configured to be responsive to one or more data read requests received from the time-series analytics application program.

Data caching for time-series analysis application

A graphics processing unit (GPU) service platform includes a control server, and a cluster of GPU servers each having one or more GPU devices. The control server receives a service request from a client system for GPU processing services, allocates multiple GPU servers within the cluster to handle GPU processing tasks specified by the service request by logically binding the allocated GPU servers, and designating one of the at least two GPU servers as a master server, and send connection information to the client system to enable the client system to connect to the master server. The master GPU server receives a block of GPU program code transmitted from the client system, which is associated with the GPU processing tasks specified by the service request, processes the block of GPU program code using the GPU devices of the logically bound GPU servers, and returns processing results to the client system.

Managing access to a resource pool of graphics processing units under fine grain control

A graphics processing unit (GPU) service platform includes a control server, and a cluster of GPU servers each having one or more GPU devices. The control server receives a service request from a client system for GPU processing services, allocates multiple GPU servers nodes within the cluster to handle GPU processing tasks specified by the service request by logically binding the allocated GPU server nodes, and designating one of the at least two GPU servers as a master server, and send connection information to the client system to enable the client system to connect to the master server. The master GPU server node receives a block of GPU program code transmitted from the client system, which is associated with the GPU processing tasks specified by the service request, processes the block of GPU program code using the GPU devices of the logically bound GPU servers, and returns processing results to the client system.

A method implemented by a server enables sharing of GPU resources by multiple clients. The server receives a request from a first client for GPU services. The request includes a first block of GPU code of an application executing on the first client. A first task corresponding to the first block of GPU code is enqueued in a task queue. The task queue includes a second task that corresponds to a second block of GPU code of an application executing on a second client. The server schedules a time for executing the first task using a GPU device that is assigned to the first client, and dispatches the first task to a GPU worker process to execute the first task at the scheduled time using the GPU device. The GPU device is shared, either temporally or spatially, by the first and second clients for executing the first and second tasks.

Queue-based GPU virtualization and management system

Embodiments are directed to methods and apparatus for making available to at least one scanning tool, information about at least one data unit shared among multiple storage objects of a plurality of storage objects stored on a storage system. The at least one scanning tool can use the information to influence at least one scanning operation on the at least some of the plurality of storage objects. Embodiments may be implemented in a computer system comprising at least one application program, at least one mapping layer that makes available to the at least one application program a plurality of storage objects, and a storage system that stores data in each of the plurality of storage objects in one or more data units. The at least one computer system further comprises the at least one scanning tool that performs the at least one scanning operation on at least some of the plurality of storage objects, and at least one redundant data elimination tool that eliminates at least some redundant data by sharing at least one of the data units among multiple storage objects.

Method for using redundant data elimination to accelerate storage system scanning

Systems and methods are provided for adopting and controlling storage resources of a distributed file system using an acquired namespace metadata service. For example, a computing system includes a first file system, and a distributed file system, which is separate from the first file system. The distributed file system includes storage nodes for storing data. The first file system includes an acquired namespace metadata server that is configured to execute on one or more nodes of the first file system. To adopt and control storage resources of the distributed file system, the first file system acquires a namespace of the distributed file system and uses the acquired namespace metadata server to manage the acquired namespace of the distributed file system. Moreover, the first file system uses the acquired namespace metadata server to directly communicate with and control access to the storage nodes of the distributed file system.

Growing community of inventors

Milford, MA, United States of America

John T Cardente