Zadara Storage, Inc.

Zadara, Inc.

Alex Lyakas

Shyam Kaushik V

Vladimir Popovski

Nelson Nahum

Yoav Ossia

Moshe Melnikov

Yair Hershko

Lev Vainblat

Shyam Kaushik, V

A method is provided for a key-value (KV) store database management system (DBMS) to provide a KV store to an application. The method includes providing a first low-level KV store and a top-level KV store. The first low-level KV store includes a root node and leaf nodes. The top-level KV store includes a leaf node pointing to the first low-level KV store. The leaf node in the top-level KV store includes a minimum key in the first low-level KV store and a pointer to the root node of the first low-level KV store. The method further includes determining a key-value insertion to the KV store would increase a number of keys in the first low-level KV store over an upper threshold and, in response to said determining, creating a second low-level KV store. The method additionally includes determining if a total number of keys in the first low-level store and the second low-level store is less than a lower threshold from key-value deletions and, when the total number of keys in the first low-level store and the second low-level store is less than the lower threshold, merging the first and the second low-level stores.

Maintaining shards in KV store with dynamic key range

A method for a storage system reduces data movement for snapshot copy-on-write operations. The method includes providing a live volume having snapshot chunks. Each snapshot chunk includes a number of deduplication chunks. Each deduplication chunk is mapped to a disk chunk on a disk. The method includes creating a snapshot of the live volume and, for each host write to a snapshot chunk of the live volume that shares a set of disk chunks with a snapshot, performing a snapshot copy-on-write operation by a using a deduplication operation.

Dedupe as an infrastructure to avoid data movement for snapshot copy-on-writes

A method is provided for a key-value (KV) store database management system (DBMS) to provide a KV store to an application. The method includes providing a low-level KV store and a top-level KV store each having its own distinct transaction. The low-level KV store includes key-value pairs. The top-level KV store includes information leading to the low-level KV store. The information in the top-level KV store includes a minimum key in the low-level KV store, a last committed transaction for the low-level KV store, and a pointer to a root node of the low-level KV store. The method further includes committing a current transaction for the low-level KV store by persisting changes during the current transaction for the low-level KV store to secondary memory, updating the information of the top-level KV store with the current transaction for the low-level KV store as the last committed transaction for the low-level KV store, and opening a new transaction for the low-level KV store.

Multiple transactions in a single KV store

A method is provided for a transaction manager in a database system to provide consistency across multiple persistent key-value (KV) stores, each having its own non-persistent transactions, for modifications to the persistent KV stores that must all persist or none at all. The method includes joining a first KV store's transaction and joining a second KV store's transaction. For a set of modifications including a first modification to the first KV store and a second modification to the second KV store that must all persist or none at all, the method includes inserting the first modification into the first KV store's transaction, inserting the second modification into the second KV store's transaction, and creating and writing a combined journal entry including first and second sub-entries describing the first and the second modifications in a persistent global journal, respectively.

Consistency across key value stores with shared journals

A method for providing virtual private storage array (VPSA) service for cloud users over a computer network includes receiving parameters for the VPSA over the network and creating the VPSA from resources of server computers. Creating the VPSA includes allocating and exposing drives that meets or exceeds specified drive characteristics, drive quantity, and array redundancy criteria to virtual controllers (VCs) in the VPSA, and dedicating parts of processor/memory complexes that each meets or exceeds a specified virtual controller hardware model to the VCs. The VCs run on virtual machines on the dedicated parts of processor/memory complexes on independent server computers. The VCs discover the exposed drives, create a virtual pool from the exposed virtual drives, implement data protection on the virtual pool, create volumes from the virtual pool, expose the volumes over the network to a customer computer, and handle access requests to the volumes from the customer computer.

Growing community of inventors

Haifa, Israel

Alex Lyakas