EXFO FTB/IQS-85100G Packet Blazer for FTB-500/IQS-600 - Manual (Page 393)

Glossary
Packet Blazer 383
Generic Framing Procedure (GFP)
GFP Mapping
Two types of mapping are currently available for GFP: framed-mapped
(GFP-F) and transparent-mapped (GFP-T), whose mappings keep the
same basic frame structure, as will be shown in the next sections. The
decision on which mode to use is dependent on the underlying service to
be transported.

Frame-Mapped GFP (GFP-F): mapping mechanism in which one
client signal frame is received and mapped in its entirety into one GFP
frame. Therefore, with this adaptation mode, the GFP-F frame size is
variable as it is directly related to the incoming client payload. In fact,
with GFP-F, the entire client frame must be buffered in order to
determine its length. GFP-F is usually used to support Layer 2 frames
like Ethernet MAC that are tolerant to some latency. The ITU G.7041
defines the following frame-mapped user payloads supported through
GFP-F:

Frame-Mapped Ethernet

Frame-Mapped PPP

Frame-Mapped Multiple Access Protocol over SDH (MAPOS)

Frame-Mapped IEEE 802.17 Resilient Packet Ring

Frame-Mapped Fibre Channel FC-BBW

Transparent-Mapped GFP (GFP-T): mapping mechanism that
facilitates the transport of 8B/10B block-coded client signals like
Gigabit Ethernet (GbE), Fibre Channel, ESCON, FICON, and DVB-ASI.
With GFP-T, individual characters of a client signal are decoded from
the client signal and then mapped into fixed-size GFP frames (64B/65B
coded superblocks). This approach avoids the buffering of an entire
client frame before it can be mapped into a GFP frame, which reduces
latency and in turn makes it ideally suited for SAN applications that
require very low transmission latency.