In Introduction Of PeerLink
Peer-based DFSR plus-100 technology, PeerLink distributed team to provide a fast and efficient file sharing collaboration. PeerLink integrated enterprise-class real-time synchronization with the distributed file locking engine. Real-time synchronization engine ensures the participation of all servers exist in the same data, no matter where that may change. This prevents users from accessing file locking member is currently being produced by another user files in any location.
PeerLink does NOT use any type of database for storage or replication. Instead, it works directly with your existing file system. One or more file collaboration jobs can be created across the PeerLink environment to work with different groupings of projects, data, servers, and/or sites. Each job consists of two or more participating servers and a folder structure on each participating server. This folder structure is called the “Watch Set” and is kept synchronized across all participating servers in real-time. In addition, locks are propagated across all participating servers as users open and begin modifying data at any one location.
The main pieces of the architecture are the PeerLink Hub/Broker and the PeerLink Agent. The former handles communication, monitoring, and management. The latter is installed on all file servers that are to participate in collaboration. In the case of NetApp environments, the PeerLink Agent is installed on a Windows machine in front of the NetApp device, allowing PeerLink to interact with NetApp.
PeerLink also includes a fully functioning web client along with the standard-rich client. This web client not only provides remote access to PeerLink’s central management console but also provides customers with a role-based access system through a local user database or Microsoft Active Directory. Out of the box, it supports admin-type accounts with full management and monitoring access and helpdesk-type accounts with the ability to remotely resolve conflicts across all collaboration jobs.
Sync-on-Save
Traditionally, a change to a file is replicated from one participating server to another only after the user has saved his or her changes to the file and has closed it. However, this does not help users who work on large files, typically without closing them for hours or even days.
With Sync-on-Save functionality enabled, PeerLink will update the other participating servers every time a user saves a file, even when the file remains open.
PeerLink leverages byte-level replication to only synchronize the parts of a file that have changed. The parts of a file that have changed are sent to the PeerLink hub, which then relays the changes to all other participating servers. The combination of byte-level replication and the hub-and-spoke architecture minimize the amount of bytes that flow between locations, keeping connections between locations as open as possible while potentially reducing costs.
The diagram below illustrates how PeerLink works within a cross-platform environment of two Windows file servers and two NetApp devices. If you use a NetApp FAS device, the PeerLink Agent will be installed on a Windows server (physical or virtual) in front of the NetApp device, communicating with it via CIFS and fPolicy. User traffic is NOT routed through the Agent server. All users continue to access the NetApp device as usual.
