International Business Machines Corporation

Iguazio Systems Ltd.

Globalfoundries Inc.

Globalfoundries U.S. 2 LLC

Orit Nissan-Messing

Daniel Isaac Goodman

Oren S Li-On

Eyal Perek

Rivka Mayraz Matosevich

Ori Modai

Yaron Haviv

Eliyahu Weissbrem

Michael Keller

Ran Harel

Yossi Siles

Aviad Zlotnick

Haim Helman

Yakov Jacob Broido

Dani Shemesh

Leonid Kull

Guy Rozendorn

Itzhack Goldberg

Assaf Nitzan

Michael Edward Factor

Thomas Charles Jarvis

Moriel Lechtman

Amichai Schreiber

Ortal Levi

Assaf Yaari

Ayelet Shani

Jonathan Amit

Robert Akira Kubo

Robert Bruce Nicholson

William James Scales

Kenneth Wayne Boyd

Chaim Koifman

Sergey Marenkov

Shachar Fienblit

Olympia Gluck

Ehood Garmiza

Kenneth F Day, Iii

Gary Valentin

John Earle Lindley

Jonathan Fischer-Toubol

Shira Ben-Dor

Amir Kredi

Lior Shlomov

Ron S Shapiro

Elena Yerushalmi

Harry Butterworth

Adam Wolman

Oren Li-on

Systems and methods are disclosed for stateless stream handling and resharding. In one implementation, a first shard including one or more messages is generated. The first shard is associated with a first shard version attribute. The first shard and the first shard version attribute are provided as a first atomic update within a data stream. The first shard is resharded into at least a second shard. The second shard is associated with a second shard version attribute. The second shard and the second shard version attribute are provided as a second atomic update within the data stream.

Stateless stream handling and resharding

Systems and methods are disclosed for dynamic application mobilization. In one implementation, an application is modeled as a processing function. The processing function is placed at a first processing node within a processing network. One or more optimization operations are computed with respect to the processing function. The processing function is mobilized from the first processing node to a second processing node within the processing network in accordance with the one or more optimization operations.

Dynamic application mobilization

Systems and methods are disclosed for stateless stream handling and resharding. In one implementation, a first shard comprising one or more messages is generated. The first shard is associated with a first state attribute. The first shard and the first state attribute are provided as an update within a data stream. In another implementation, a first shard including a first state attribute is received within a first stream. A message that is inconsistent with the first state attribute is identified within the first shard. The message is associated as an attribute of the first shard. A second shard including a second state attribute is received. Based on the second state attribute, a position of the message within the second shard is determined. The message is inserted into the second shard based on the determining.

Systems, methods, and related technologies for application-centric object storage and configuration are described. In certain aspects, a processing device can receive, from an application, one or more requests with respect to a data object. The processing device can process the one or more requests to determine one or more data utilization characteristics of the application with respect to the data object. The processing device can configure one or more aspects of the data object in view of the one or more data utilization characteristics.

Application-centric object configuration

A method of managing data with high granularity, comprises identifying data objects and an associated data management policy. The policy uses features and a common semantic to define a feature-based sequence of data management actions of a data management operation that varies for different objects depending on respective features. Features of the data objects are obtained and then used to associate a data management action with the object using the policy so that the object is managed individually according to its own features, thus achieving high granularity of data management precision and also high flexibility.

Service oriented data management and architecture

A method or system of managing data with high granularity, comprises identifying data objects and an associated data management policy. The policy uses features and a common semantic to define a feature-based sequence of data management actions of a data management operation that varies for different objects depending on respective features. Features of the data objects are obtained and then used to associate a data management action with the object using the policy so that the object is managed individually according to its own features, thus achieving high granularity of data management precision and also high flexibility.

System and method of managing service oriented data

Systems and methods are disclosed for structuring unstructured machine-generated content. In one implementation, unstructured content is received and processed to identify a first content segment. Parameter(s) within the first content segment are identified and classified. A content segment skeleton that reflects an arrangement of the first content segment is generated. Based on the classifying of the parameter(s), the parameter(s) are extracted s from the first content segment into a structured content element in a structured content format, with the structured content element including a representation of the content segment skeleton. Based on the structured format, a query adapter is generated. Queries are executed via the query adapter and the structured format.

Structuring unstructured machine-generated content

Systems, methods, and related technologies for application-centric object storage and managing object requests via multiple indexes are described. In certain aspects, a processing device can receive a first data object, generate, with respect to the first data object, a first index, generate, with respect to the first data object, a second index through which the first data object can be accessed, receive a request with respect to the first data object, process the request to determine which of the first index or the second index to utilize in completing the request, and complete the request via at least one of the first index or the second index.

Managing object requests via multiple indexes

Systems, methods, and related technologies for application-centric object storage and interfaces. In certain aspects, a processing device receives, from a first application, one or more requests with respect to a data object. The processing device processes the one or more requests to identify one or more substructures of the data object. Based on an identification of the one or more substructures of the data object, the processing device defines a storage structure for the data object. The processing device defines an interface through which the first application can provide data access commands to the data object as stored in the defined storage structure.

Application-centric object interfaces

Methods, apparatus and computer program products implement embodiments of the present invention that include storing one or more data volumes to a small computer system interface storage device, and receiving a request to map a given data volume to a host computer. One or more attributes of the given data volume are identified, and using the identified one or more attributes, a unique logical unit number (LUN) for the given data volume is generated. The given data volume is mapped to the host computer via the unique LUN. In some embodiments, the generated LUN includes one of the one or more attributes. In additional embodiments, the generated LUN includes a result of a hash function using the one or more attributes. In storage virtualization environments, the data volume may include secondary logical units, and mapping the given data volume to the host may include binding the SLU to the host.

Using deterministic logical unit numbers to dynamically map data volumes

Systems, methods, and related technologies for application-centric object storage and virtual data objects are described. In certain aspects, a processing device can process a first data object in a first format to generate a virtual data object, the virtual data object including a representation of the first data object in a second format. The processing device can receive a request with respect to the first data object from an application that is configured with respect to the second format. The processing device can complete the request with respect to the first data object and the application via the virtual data object.

Virtual data objects

Methods, computing systems and computer program products implement embodiments of the present invention that include configuring multiple storage system resources to manage a storage entity distributed among the storage system resources. Upon receiving, by a first given storage system resource from a host computer, an I/O request for data in the storage entity, a second given storage system resource responsible for managing the storage entity data referenced in the I/O request is identified, and the I/O request is forwarded to the second given storage system resource. Upon receiving the forwarded I/O request, the second given storage system resource performs a data compression operation while processing the I/O request, and conveys a result of the I/O operation to the first given storage system resource. Upon receiving the result from the second given storage system resource, the first given storage system resource forwards the result to the host computer.

Data compression for grid-oriented storage systems

Methods, apparatus and computer program products implement embodiments of the present invention that include receiving, by a first computer in communication with a storage system, multiple resource packs, each of the resource packs including a plurality of attributes for storage resources on the storage system. Upon the first computer receiving a request from a second computer for a storage service, one or more of the received resource packs are identified that have respective attributes matching the storage service, a given identified resource pack is selected and the given identified resource pack is conveyed to the second computer. In some embodiments, the given identified resource pack can be conveyed by conveying an allocation request to the storage system for a portion of the given identified resource pack, and in response to the request, the storage system can allocate the requested portion of the given identified resource pack to the second computer.

Matching storage resource packs to storage services

Growing community of inventors

Hod Hasharon, Israel

Orit Nissan-Messing