Cohesity, Inc.

Sharath Kumar Naeni

Sachin Jain

Venkatesh Pallipadi

Anubhav Gupta

Ganesha Shanmuganathan

Zhihuan Qiu

Anirvan Duttagupta

One or more buckets of key-value pairs of a first node of a distributed storage system are selected to be migrated to a second node of the distributed storage system. One or more underlying database files corresponding to the one or more selected buckets are identified. The one or more identified underlying database files are directly copied from a storage of a first node to a storage of the second node. The copied underlying database files are linked in a database of the second node to implement the one or more selected buckets in the second node.

Fast migration of metadata

A method and system for managing a storage cluster is described. An identification of a plurality of nodes of a storage cluster to be removed from the storage cluster is received. It is determined whether to allow the removal of the plurality of nodes. In response to a determination that the removal of the plurality of nodes is allowed, identifications are made of data migrations to be performed to rebalance the data among nodes remaining in the storage cluster after removal of the plurality of nodes. The data migrations are performed. The plurality of nodes are removed from the storage cluster.

Multi-node removal

Growing community of inventors

Santa Clara, CA, United States of America

Sharath Kumar Naeni