Deduplication concept and models
Deduplication is a feature that
-
removes duplicate data to ensure only unique data packets are processed and stored
-
preventss unnecessary storage consumption by eliminating redundant copies of traffic, and
-
supports two operational models.
Due to the continuous rise in data flow, navigating in the networks and the data processed in different applications leads to data duplication both in source and target. For efficient data management, security, and storage you can use deduplication to remove duplicate data.
The deduplication feature eliminates the duplicate traffic which is received from traffic analysers or data storage tools. It identifies duplicate flows which are transferred in the Nexus Data Broker (NDB) switch.
Deduplication feature can be used to identify and eliminate duplicate data flows as they traverse data broker switches. By removing redundant copies of traffic—often generated by traffic analyzers or data storage tools—deduplication ensures that only unique data packets are processed and stored.
Deduplication supports two model types:
-
Inline deduplication Model (Interface Mode) : You can configure the inline model on existing NDB switches. This eliminates duplicate packets using interfaces. Initial interface packets are forwarded and the duplicate packets from other interfaces are dropped.
-
Deduplication on a Stick Model (VLAN Mode) : In this model, the packet flow is bound to a specific VLAN. The packet flow is permitted on the first VLAN, and duplicate flows on other VLANs are restricted. The packets are tagged with specific VLAN when the flow enters the data broker switches. Each packet is tagged with a unique VLAN, the packets with VLAN tag are forwarded to the deduplication switch, and duplicate flow is removed on that switch.
Feedback