Datadirect Networks, Inc.

Rajkumar Joshi

David F Fellinger

Jean-Luc Chatelain

Michael Piszczek

Dan Olster

Dave Fellinger

Systems and methods for dynamic context based data protection and distribution are described herein. In the method data replication in a multi-zone storage system automatically changes based on historical storage trends and/or external information obtained from third party information sources, such as feeds, websites, news alerts, emergency alerts, weather alerts and databases. The methods include monitoring third party information sources and evaluating a correlation between event data from the third party information sources and key words that are user provided and/or system derived. A correlation between the event data and data storage history is evaluated. A storage policy is derived based on the results of the correlation evaluation(s). The derived storage policy and corresponding key words are stored in a database. Data is stored in storage arrays according to storage policies corresponding to the key words identified in the data.

Dynamic data protection and distribution responsive to external information sources

The present invention is directed to a distributed data migration and storage system and method configured for fully parallel writing and retrieval of data objects to (and from) the Web Object Scaler (WOS) object storage system with highly efficient accessibility and using a simplified set of filing commands. The data objects migrate between data generating entities and the WOS object storage through an intermediate Burst Buffer tier (supported by the Infinite Memory Engine) without the utilization of a serial process substantially simultaneously through parallel channels. The metadata for each object is captured with appropriate pointers for efficient finding of the object in the WOS storage system. The WOS nodes return the stored objects' ID (OIDs) corresponding to the objects storage address which are used for the object access.

Method and system for parallel file operation in distributed data storage system with mixed types of storage media

Systems and methods for automated firmware update with rollback are described herein. The systems include a plurality of storage zones, each storage zone including a plurality of storage nodes, each storage node including a plurality of storage media. The method includes monitoring storage system activity and parameters and maintaining a data storage system usage and parameter database containing system activity information. When a firmware update is available, data storage system activity is evaluated. Storage nodes needing the firmware update are identified. The firmware update is run on available storage nodes identified as needing the firmware update. The impact of the firmware update is evaluated and a rollback of the firmware update is initiated on all firmware updated storage nodes when parameter variations are significant and/or result in degraded performance.

Automated firmware update with rollback in a data storage system

Systems and methods for storing data in a data storage system that allows dynamic context-based data protection and distribution are disclosed. The method includes receiving a storage request and evaluating whether a storage policy is specified in the storage request. When the storage request specifies a storage policy, the data is stored according to the specified policy. When the storage request does not specify a storage policy, meta data and/or the data item itself may be evaluated to identify a storage policy. When a storage policy has been identified, store the data is stored according to the identified policy. When a storage policy has not been specified and cannot be identified, an error is returned.

Dynamic context-based data protection and distribution

Systems and methods for adaptive asynchronous data replication in a data storage system are described herein. The data storage system includes a plurality of zones each having a plurality of storage nodes, each having a plurality of storage devices. The system provides for replication according to policies associated with data items such that data items are stored among a plurality of zones. The data items are stored as one or more objects and may be replicated asynchronously or synchronously. The system adapts the synchronicity of the replication based on a combination of criteria including a put performance threshold, a backlog threshold, latency and the current synchronicity status. By automatically adjusting the replication synchronicity, the system and methods achieve improved performance while maintaining data resiliency.

Growing community of inventors

Simi Valley, CA, United States of America

Rajkumar Joshi