Abstract: Implementations are provided herein for establishing multi-protocol redirection links that are associated with an object, a file, a set of files, directories, etc. The multi-protocol redirection link can transparently redirect a client to a different part of the namespace or even to a different storage cluster to access data. For example, using multi-protocol redirection links, a single namespace in a cluster of nodes can be scaled up with multiple secondary storage clusters being the eventual redirect path. A multi-protocol redirection table can be established as a basis for the multi-protocol redirection link. The file system can be aware of discriminative information associated the client, for example, the network protocol the client is using, the access zone the client is connecting to, user identification of the client, group identification of the client, geography of the client, etc.

Abstract: Implementations are provided herein for dynamic ephemeral point-in-time snapshots being taken upon read requests from clients associated with object protocols and/or other protocols that depend on a consistent read. In response to receiving a read request from a protocol that depends on a consistent read, a point-in-time snapshot can be taken on the read target that maintains the durability of the point-in-time snapshot, in memory, until the client no longer requires access to the file. It can be appreciated that prior to taking the snapshot, in contemporaneously with taking the snapshot, or after the snapshot is taken, a Network File System (“NFS”) client, a Sever Message Block (“SMB”) client, or other clients of network protocols that allow for modifications and overwrites of file data, can open and modify the file that is the basis of the point-in-time snapshot, without repercussion to the object protocol that is using the point-in-time snapshot version of the file for a consistent read view.

Abstract: Implementations are provided for scalable cloud backup. A coordinator process can manage worker processes on nodes to package file system data that is targeted for cloud backup into node local upload objects. File data can be arranged into distinct block offsets of the node local upload object. A set of metadata tables can be generated that characterize each file that is backed up as well as file block location information for each data block. The node local upload objects can be uploaded to a cloud service provider. The set of metadata tables generated by the worker process can be coalesced into a global set of metadata tables that describe the data that has been backed up. In one implementation, after an initial cloud backup has occurred, a snapshot service of the file system can be used to incrementally backup blocks of the file that have been changed.