Emc IP Holding Company LLC

Emc Corporation

Tal Ben-Moshe

Niko Farhi

David Krakov

Eli Dorfman

Kirill Shoikhet

Renen Hallak

David Meiri

Zvi Schneider

Roman Vainbrand

Noa Cohen

Lior Kamran

Kobi Luz

Yaron Segev

Vladimir Shveidel

Xiangping Chen

Anton Kucherov

Amitai Alkalay

Uri Shabi

Ronen Gazit

Philip C Love

Yuval Harduf

Ido Halevi

Gilad Braunschvig

Irit Yadin Lempel

Vladislav Weinbaum

A method comprising: storing, in a memory, a mapping tree that is implemented by using an array of mapping pages, the mapping tree having a depth of D, wherein D is an integer greater than or equal to 0; receiving a write request that is associated with a first type-1 address; storing, in a storage device, data associated with the write request, the data associated with the write request being stored in the storage device based on a first type-2 address; generating a map entry that maps the first type-1 address to the first type-2 address; calculating a first hash digest of the first type-1 address; and storing the map entry in a first mapping page.

Hash-based one-level mapping for storage clusters

Described herein are embodiments of a process for efficiently allocating RAID stripes across an array of disks (e.g., SSDs). In some embodiments, the process can be used to allocate RAID stripes across a 'heterogeneous' storage array (i.e., an array of different sized disks). Also described herein are embodiments of a storage system that utilize said processing.

Data protection in a heterogeneous random access storage array

A System, computer program product, and computer-executable method of transitioning replication on a data storage system including a production site and a replication site from asynchronous replication to synchronous replication, the System, computer program product, and computer-executable method including receiving a request to transition from asynchronous replication to synchronous replication and changing the data storage system to a transition mode, wherein the data storage system transitions the data storage system from using asynchronous replication to synchronous replication.

Consistent transition from asynchronous to synchronous replication in hash-based storage systems

A hiccup management scheme for use within a storage system can maintain low latency on client I/O when a storage device is temporarily unavailable. In some embodiments, a storage using uses double parity data protection can tolerate concurrent hiccups by up to two storage devices within a storage array.

Hiccup management in a storage array

A computer program product, system, and method for generating coded fragments comprises receiving, at a first node of a distributed storage system, an I/O request to write a block; splitting the block into a plurality of sub blocks; generating a sub block error detection hash for each of the sub blocks; sending the block and the sub block error detection hashes to a second node within the distributed storage system; and storing the block and the sub block error detection hashes to a storage device at the second node.

End-to-end data protection for distributed storage

Described herein are embodiments of a process that can be used to balance the allocation of primary memory between different types of information. In some embodiments, the memory allocation is balanced dynamically based on observed I/O patterns. Related system embodiments are also described.

Memory allocation balancing for storage systems

Described embodiments may provide systems and processes for migrating a replica hierarchy from at least one source device to at least one target device in a storage system. A configuration of the replica hierarchy is migrated, one or more internal nodes of the replica hierarchy are migrated, and one or more leaf nodes of the replica hierarchy are migrated. A cutover operation is performed by deactivating the at least one source device associated with the included replica hierarchy.

Live migration of a tree of replicas in a storage system

A system, computer program product, and computer-executable method of managing parity data of a Redundant Array of Independent Disks (RAID) on a data storage system, where the data storage system stores a first set of parity data and retains metadata associated with the first set of parity data, the system, computer program product, and computer-executable method including allocating a set of locations on the data storage system for a second set of parity data, processing the second set of parity data, and updating the metadata associated with the first set of parity data.

Dynamic parities mechanism

A system, computer program product, and computer-executable method of implementing a redundant array of independent disk (RAID) system wherein the RAID, the computer-executable method comprises storing data storage blocks arranged in a first plurality of data rows and a second plurality of data columns and storing parity data in defined parity blocks, wherein a portion of the defined parity blocks include column parity data.

Updated raid 6 implementation

Described herein are several embodiments of systems and processes to decrease startup time for subsystems of a storage system. According to some embodiments, subsystem memory is allocated using memory-mapped files.

Fast startup for modular storage systems

In one aspect, a method includes providing a server that stores a hash digest metadata table and configuring entries in the hash digest metadata table to include entries that include persistent fields and non-persistent fields. The hash digest metadata table maps a hash digest representing X-Page data to its metadata information. The persistent fields include a short hash digest field, a physical layout offset field, a deduplication reference count field and a flags field.

Data structure for hash digest metadata component

In one aspect, a method includes sending a short hash handle from a production site to a replication site, the short hash handle being generated using a hash signature and being smaller than the hash signature, determining at the replication site whether the short hash handle exists at the replication, reading and sending the hash signature associated with the short hash handle if the short hash handle exists at the replication site, determining at the replication site whether the hash signature exists at the replication and emulating a write operation by associating the hash signature to an address in a second storage volume at the replication site if the hash signature exists at the replication site. The hash signature is generated using a data block from a first storage volume at the production site.

Hash-based replication

In one aspect, a method includes receiving and recording journal data at a client on a first server, providing the data to a journal component, transferring the journal data from the journal component to a reserve memory in a memory of a second server using a remote direct memory access (RDMA), after failure of the client, retrieving the journal data from the reserve memory using the RDMA and replacing the journal data lost from the failure of the client with the journal data from the reserve memory.

Backing up journal data to a memory of another node

In one aspect, a redundant array of independent disk (RAID) memory storage system includes a plurality of data storage disks comprising data blocks, a first redundancy storage disk comprising parity blocks that are a parity by row of the data blocks in the data storage disks, a second redundancy storage disk comprising parity blocks storing parities of diagonals of the data blocks in the data storage disks, a third redundancy storage disk comprising parity blocks storing parities of diagonals of the data blocks in the data storage disks and circuitry configured to recover data from three failed storage disks.

Growing community of inventors

Kiryat Ono, Israel

Tal Ben-Moshe