Emc IP Holding Company LLC

Emc Corporation

Emc International Company

Alex Solan

Jehuda Shemer

Assaf Natanzon

Saar Cohen

Steven Richard Bromling

Yossef Saad

Leehod Baruch

Gabi Benhanokh

Nir Shasha

Valerie Lotosh

Erez Sharvit

Lev Ayzenberg

Ido Singer

Oded Kedem

Tzach Schechner

Ron Cooper

Asaf Natanzon

Jehuda Shemer (udi)

Ron Pantofaro

Avi Owshanko

Tomer Twig

Asaf Bar'el

Data protection operations based on direct storage access. Data protection operations that involve large data transfers are optimized or improved by transferring the data using a communication path that includes direct access to disks of a storage array. This avoids latencies associated with transferring data through the layers of the storage array. The locations of the data to be transferred are identified and provided to an appliance. The appliance can then read and transfer the data over a communication path that includes direct disk access.

Fast recovery in recoverpoint using direct storage access

Described is a system for intelligent and reversible data masking of computing environment information shared with an external system. The system may leverage a secure masking agent that acts as an intermediary between a computing system (or environment) and an analytics component. The masking agent may provide real-time reversible data masking that ensures that sensitive information is not exposed outside of a secure (e.g. on-premises) environment, while at the same time ensuring the analytics component receives sufficient contextual information to perform a detailed analysis with the shared information. For example, the system may identify and mask identifying information of a particular server or host, while still retaining certain contextual information such as a network topology.

Intelligent and reversible data masking of computing environment information shared with external systems

One example method includes performing various operations while a stream of IOs is being applied to a source dataset and replicated to a target dataset, and the operations include receiving a replicated IO that was previously applied to a source dataset Merkle tree, applying the replicated IO to a target dataset Merkle tree by entering an updated hash value in the target dataset Merkle tree, and applying a timestamp to the target dataset Merkle tree.

Using merkle trees in any point in time replication

Replication operations of replicating data from a production site to a replica site. The replication is performed using independent Merkle trees. The Merkle trees are updated asynchronously using Merkle trees that have been augmented with a time-based value. The synchronization is verified by comprising root hashes of the independent Merkle trees at certain points in time. The replication and Merkle trees are self-healing and trigger a resynchronization when a discrepancy is discovered.

Self healing fast sync any point in time replication systems using augmented Merkle trees

One example method includes creating a source FS Merkle tree having a structure that matches a topology of a source FS, and a target FS Merkle tree having a structure that matches a topology of a target FS, receiving a change to the source FS, implementing the change in the source FS, updating the source FS Merkle tree to reflect the change, replicating the change to the target FS so that, with respect to the change, the target FS is synchronized with the source FS, updating the target FS Merkle tree to reflect the change to the target FS, and comparing the source FS Merkle tree to the target FS Merkle tree.

Filesystem embedded Merkle trees

One example method, which may be performed in connection with an object store, includes receiving a key of a key-value pair, correlating the key to a location in a base of a Merkle tree, inserting the key at the location, hashing the value associated with the key to produce a data hash, and inserting the data hash in the Merkle tree. The Merkle tree may then be checked for consistency, and synchronized with another Merkle tree. The Merkle tree may be of a fixed size, and insertion of the key in the Merkle tree does not change the location of any keys existing in the Merkle tree prior to insertion of the new key.

Using sparse merkle trees for smart synchronization of S3

A method, apparatus, and system for transmitting file system metadata from an indexing splitter running in a VM to a source side RPA is disclosed. The operations comprise: capturing one or more file system events in a production virtual machine (VM) at an indexing splitter; transmitting file system metadata representing the captured file system events from the indexing splitter to a data splitter, the data splitter being an agent running on a host system hosting the VM; transmitting the file system metadata inside one or more special input/output (I/O) commands associated with a predetermined tag from the data splitter to a source side replication protection appliance (RPA) alongside regular storage system I/O command data; identifying the special I/O commands at the source side RPA based on the predetermined tag; and recovering the file system metadata from the special I/O commands at the source side RPA.

Indexing splitter for any pit replication

An agent executing on a processor at a data production site detects filesystem events and creates a list of bookmarks describing the filesystem events in user-understandable strings. Each bookmark is associated with metadata about the filesystem event that includes a timestamp and a point-in-time (PiT) snapshot of the filesystem at the time of occurrence of the event. The bookmarks and associated metadata are transferred to a recovery site and the bookmark set is recorded in storage. The storage can be searched for bookmarks in the set that reference the filename of a file of interest, and the file recovered and replicated using an associated PiT.

Filesystem operation bookmarking for any point in time replication

Embodiments extend using sparse Merkle trees for smart synchronization of S3 buckets by overcoming fixed size limitations through creating another Merkle tree when the fixed size limit of the first tree is exceeded, and creating yet another tree when the second tree is filled up, and so on as needed. The method maintains a super Merkle tree of trees, in which each tree can be synchronized separately by keeping a strict division to trees according to generation number. The generation is passed from a source site to a target site during replication operations. Syncing between two data sites is efficient as the super Merkle tree of the source is synced with the super Merkle tree of the target using the hashes on the nodes, as in normal Merkle tree sync operations.

Merkle super tree for synchronizing data buckets of unlimited size in object storage systems

Embodiments extend using sparse Merkle trees for smart synchronization of S3 buckets by overcoming fixed size limitations through creating another Merkle tree when the fixed size limit of the first tree is exceeded, and creating yet another tree when the second tree is filled up, and so on as needed. The method maintains a list of trees, in which each tree can be synchronized separately by keeping a strict division to trees according to generation number. The generation is passed from a source site to a target site during replication operations. The tagging of the generation number also makes it easy and efficient to remove an older version of an element or deal with deleted elements. This allows efficient syncing between two data object buckets without a size limitation on number of elements in a bucket.

Merkle tree forest for synchronizing data buckets of unlimited size in object storage systems

Described is a system for intelligently distributing the retrieval of recovery data amongst peer-based and cloud-based storage sources. The system may leverage a peer network by performing a recovery from a combination of sources including peer devices and a cloud-based storage by considering a multitude of factors. For example, the system may consider retrieval information (e.g. availability, performance metrics, retrieval costs, etc.) associated with each of the potential sources in combination with target information (e.g. recovery objectives, file recovery priorities, etc.) to create an ordering preference (e.g. ranking) for sources from which to retrieve recovery data. The system may rank the sources by determining a specialized recovery score for each source thereby allowing the intelligent retrieval of recovery data.

Intelligently distributing retrieval of recovery data amongst peer-based and cloud-based storage sources

Described is a system for intelligently storing and recovering backup data using multiple storage tiers of a cloud-based storage. The intelligent storing of backup data may leverage different cost structures associated with various cloud-based storage tiers by efficiently distributing data to more cost effective tiers. The system may also provide a coordinated (e.g. phased) recovery that prioritizes the recovery of certain files so that a user can gain access to such files in a shorter amount of time. The system may also provide enhanced storage and recovery functionality (e.g. via specialized user interface elements) by leveraging placeholder files to offload and recover individual files, and expedite the recovery of individual files during a substantial recovery process.

Growing community of inventors

Hertzliya, Israel

Alex Solan