Emc IP Holding Company LLC

Dell Products Inc.

Emc Corporation

Emc International Company

Jehuda Shemer

Assaf Natanzon

Kfir Wolfson

Saar Cohen

Amit Lieberman

Leehod Baruch

Ron Bigman

Alex Solan

Valerie Lotosh

Ofir Ezrielev

Erez Sharvit

Jawad Said

Stav Sapir

Naor Radami

Arieh Don

Boris Shpilyuck

Itay Azaria

Amos Zamir

Adam E Brenner

Amihai Savir

Ravi Vijayakumar Chitloor

Igor Dubrovsky

Lev Ayzenberg

Slavik Neymer

Steven Sadhwani

Bing Liu

Steven Richard Bromling

Oded Peer

Yana Vaisman

Tomer Kushnir

Maxim Drobachevsky

Nadav Azaria

Anand Rudrabhatla

Ronen Rabani

Shahar Frank

Oron Golan

Ran Goldschmidt

Zion Drori

Ilan Buyum

Ido Begun

George Mathew

Abhinav Duggal

Mark D Malamut

Yossef Saad

Arun Murti

Krishna Deepak Nuthakki

Gabi Benhanokh

Chris E Rowen

Srinivas Kangyampeta

Aviram Fireberger

Ron Cooper

Sharon Vitek

Alexandra Solan

Anton Kucherov

Avitan Gefen

Sumedh Wasudeo Sathaye

Dennis Thomas Duprey

Karl M Owen

Felix Shvaiger

Mahesh Kamat

Yaron Dar

Jennifer M Minarik

Ido Singer

Naveen Rastogi

Chegu Vinod

Reut Kovetz

Kelly L Lisai

Patrick East

Matan Gilat

Georges Brun-Cottan

Meir Pinhasov

Prasanna Malaiyandi

Jonathan Grey

Lehood Baruch

Yuval Nir

Peleg Yiftachel

Methods and systems for managing inference models are disclosed. An inference model may be divided into any number of sub-network units and each sub-network unit of the sub-network units may be located in a different geographical location. Data privacy regulations may apply to the geographical locations. Therefore, a first sub-network unit may be subject to a first data privacy regulation and a third sub-network unit may be subject to a second data privacy regulation. If an output of a sub-network unit is not usable by a downstream sub-network unit due to differences in data privacy regulations, the first sub-network unit may be replaced with a second sub-network unit, the second sub-network unit generating an output that does meet the data privacy regulations of the downstream sub-network unit. Doing so may reduce interruptions to inference generation by the inference model.

Managing inference models to meet data privacy regulations

A data protection system includes a splitter configured to reduce latencies when splitting writes in a computing environment. The splitter captures a write and adds metadata to augment the write with virtual related information. The augmented data is provided to a smartNIC while the write is then processed in the IO stack. The smartNIC may have a volume only visible to the splitter. The smartNIC also includes processing power that allows data protection operations to be performed at the smartNIC rather than with the processing resources of the host.

Smartnic based virtual splitter ensuring microsecond latencies

Methods and systems for authenticating data processing systems throughout a distributed environment without user intervention are disclosed. To authenticate data processing systems without user intervention, a system may include a network core and one or more data processing systems. Data processing systems may provide telemetry data to the network core for use in future authentication processes. To increase security throughout the system, the network core may ensure that low security data points are not retained for authentication. A low security data point may be a data point that matches a data point obtained from another data processing system. If a root of trust is lost between a data processing system and the network core, the network core may use retained telemetry data points to generate a security questionnaire. If the data processing system successfully answers security questions of the security questionnaire, the root of trust may be restored.

System and method for data retention based on commonalities between data processing systems

Compressing files is disclosed. An input, which is associated with an original file and new content, is to be compressed. The input includes a consensus sequence of the original file and the new content. The new content is aligned based using the consensus sequence of the original file in order to generate a new consensus sequence that reflects both the original content and the new content. The compression engine generates a new compression matrix and a new consensus sequence. Using the new consensus sequence, pointer pairs are generated. Each pointer pair identifies a subsequence of the consensus matrix. The new compressed file includes the pointer pairs and the new consensus sequence.

Adding content to compressed files using sequence alignment

One example method includes performing a filtering process that identifies one or more candidate hosts for scheduling of a pod, wherein the candidacy of a host is determined based in part upon an association rule, generating an overall host score for each of the candidate hosts, and scheduling the pod to one of the candidate hosts based on the overall host score of that candidate host. A host risk score and/or pod risk score may be used in the generating of the overall host score.

Kubernetes security isolation based on classified information and risk levels

Compressing files is disclosed. An input file to be compressed is first aligned. When the file has multiple axes or dimensions, the file is aligned along each of the axes. Aligning the file includes splitting the file into sequences that can be aligned along each of the axes or dimensions. Aligning the file generates a compression tensor, where each row or dimensional space of the compression tensor corresponds to part of the file. A consensus tensor is determined from the compression tensor. Using the consensus tensor, pointer lists are generated. Each pointer lists identifies a subsequence or portion of the consensus tensor. The compressed file includes the pointer lists and the consensus tensor.

Compressing multiple dimension files using sequence alignment

Providing content-based encryption to content data in a data processing system by creating datasets by grouping metadata for data objects that are intended to be encrypted with a common encryption key, where each dataset spans multiple storage devices of different storage types, and defines a single data encryption unit for the data objects referenced by a respective dataset. Each dataset is tagged with an encryption tag to enable or disable use of a self-selected encryption key. Encryption keys stored in or made available to the system are accessed for encrypting the data objects using an encryption process. A key management component maps each dataset to a corresponding encryption key of the encryption keys, and an encryption component encrypts, for each dataset, referenced data objects using a corresponding mapped encryption key.

Encryption key management using content-based datasets

Methods and systems for authenticating data processing systems throughout a distributed environment without user intervention are disclosed. To authenticate data processing systems without user intervention, a system may include a network core and one or more data processing systems. The network core may attempt to authenticate data processing systems using a security questionnaire. Security questions in the security questionnaire may be based on telemetry data obtained from the data processing system prior to a loss of a root of trust. To conserve computing resources, only telemetry data with a security score that exceeds a security score threshold may be retained. The network core may provide the data processing system with a security questionnaire and the data processing system may use similar telemetry data to respond to the security questionnaire. If the answers to the security questions are considered accurate, the data processing system may be re-authenticated.

System and method for selecting security questions for device authentication based on security scores of data

A method includes searching a group of PITs, identifying one of the PITs as having an indicator of an occurrence of an event involving data associated with the identified PIT, restoring the data, running a production system copy using the data, and while the production system copy is running, taking increasingly granular backups of the data until the event is located. The event may be a corruption of the data, or other problem.

Cyber recovery forensics kit—efficient event searching over pits

One example method includes defining an airgap control policy that specifies a threshold data value, generating a value for a set of data, determining whether the value that has been generated for the data meets or exceeds the threshold data value, and opening the air gap when the value that has been generated for the data meets or exceeds the threshold data value. The airgap is closed automatically when the value that has been generated for the data meets or exceeds the threshold data value.

On-demand operational airgap policy—value threshold

One example method includes performing delta operations to protect data. During a delta operation, a primary map and a secondary map are processed using bit logic. The bit logic determines how to handle data stored at a location on the volume associated with an entry in the primary map and included in the current delta operation when a new write for the same location is received as the corresponding entry in the primary map is processed.

Optimized one pass delta streaming filter

One example method includes receiving, at a remote site from a production site, copies of production site assets, storing, at the remote site, the copies of the production site assets, using, at the remote site, the copies of the production site assets to restore a temporary production site, running the temporary production site at the remote site, and restoring, from the remote site to the production site, the copies of the production site assets.

Recovery by multi-cloud handoff

A system can maintain a first data center that comprises a virtualized overlay network and virtualized volume identifiers, and store data comprising virtualized workloads. The system can determine a service level agreement associated with providing a second data center as a backup to the first data center. The system can, based on the service level agreement, divide, into a first portion of tasks and a second portion of tasks deploying the data to a secondary storage of the second data center, deploying the data to a primary storage of the second data center, and configuring the second data center with the virtualized overlay network and the virtualized volume identifiers. The system can perform the first portion of tasks before determining to restore the first data center to the second data center. The system can perform the second portion of tasks in response to determining to restore the first data center.

Standby data center as a service

Providing content based data protection for data stored in a large-scale data storage system by scanning data stored in one or more databases for discovery of metadata, and extracting the discovered metadata, for storage in a data catalog, the data catalog having a scanning function performing the scanning step, and comprising a database storing the metadata in one or more tables. A protection policy is defined to commonly protect content data referenced by metadata in the data catalog, and applied to the referenced content data to perform a data protection operation the content data. Datasets stored in the catalog are generated by running queries on the catalog, where a query comprises metadata selectors as tags applied to the catalog, where the tags define at least one of a file type, name, location, creation time, or file characteristic.

Data catalog for dataset lifecycle management system for content-based data protection

Methods and systems for authenticating data processing systems throughout a distributed environment without user intervention are disclosed. To authenticate data processing systems without user intervention, a system may include a network core and one or more data processing systems. A previously established root of trust between the network core and a data processing system may be lost and the network core may attempt to re-authenticate the data processing system using shared knowledge. The shared knowledge may be based on data obtained from the data processing system and an inference generated by a twin inference model. The network core may provide the data processing system with a security questionnaire based on the shared knowledge and the data processing system may use the twin inference model to respond to the security questionnaire. If the answers to the security questions are considered accurate, the data processing system may be re-authenticated.

System and method for device authentication in a distributed environment using a twin inference model

Growing community of inventors

Kfar Saba, Israel

Jehuda Shemer