Network congestion is a problem in data networks that can impact the performance of the network. Identifying the source of the congestion and mitigating it can be a difficult task.
To help address network congestion, Fabric Performance Impact Notifications (FPINs) are defined to help identify types and sources of congestion and allow the receivers of the information to act in a manner to mitigate the congestion.
The notifications are delivered through FPIN ELS (Extended Link Service) and a new congestion primitive. Determination of whether the congestion primitive is supported is determined through the negotiation with the EDC (Exchange Diagnostic Capabilities) ELS. Determination of which types of notifications are supported by the switch is negotiated with RDF (Register Diagnostic Functions) ELS. The negotiated types of notification are Link Integrity notification, Delivery notification, Peer Congestion notification and Congestion notification (if not negotiated with the EDC ELS).
To allow debug of sending and receiving FPINs, the SANBlaze solution will provide mechanisms to allow FPINs with various notifications in the payload to be sent to the Fabric Controller and specified N_PORTs. The solution will also allow receiving FPINs. Logging and tracing will be provided for any FPIN received or sent. In addition, counters will be provided for the notifications contained in the FPINs which are sent or received. The solution will also allow viewing of the notifications contained within the FPINs sent and received.
Download our FPIN Verification white paper below to learn more.
You can view all of our white papers here.