Today?s optical transport networks are evolved to provide extremely cost efficient and robust voice channel communication over larger distances. The intellectual and material investments that have been done to create these networks and their supporting management and control infrastructure have been considerable. Little consideration have been given to the transportation of any LAN, storage area networks (SAN) or any other network over these network because of the dominance of voice traffic over data traffic.?
Various data LAN and SAN protocol are evolved to satisfy different constraints. They are optimized for geographically short reach, since data traffic patterns are highly localized. Consequently no attempts are made to utilize the optical transport networks that span the globe.?
Recently data traffic has been outpaced by voice traffic and data traffic patterns are in the midst of an inversion in which local concentration are giving way to more dispersed traffic over larger areas. As a result, existing LAN, WAN and MAN data networking solutions become candidates for integration into the optical transport network, preferably under a common data framework.
Over the last few years, a need has arisen for a simple traffic adaptation mechanism, which will integrate the existing physical, and data link formats into common public transport network infrastructure. The desired traffic adaptation mechanism should be simple enough to scale gracefully with the ever-increasing transmission rate in the core of the transport networks. It must be flexible enough to accommodate diverse data transmission requirements.
Generic Framing Procedure is a simple but flexible traffic adaptation mechanism specially designed to transport either block-coded or packet-oriented data streams over a byte synchronous channel. GFP generalizes the error control based frame delineation scheme successfully employed in ATM for both fixed and variable length data. GFP relies on the length of current payload and an error check for frame boundary delineation. Successful validation of these two pieces of information, conveyed in GFP frame header, is used to determine proper data link synchronization and the number of bytes of the incoming frame.
Processing of arbitrary blocks of bytes at a time substantially reduces the processing requirements for data link mapper/demapper. Exploiting the low bit error rate performance of fiber based communication media for data link synchronization logic decreases the receiver logic. This reduced implementation complexity makes GFP particularly suitable for high speed transmission links such as point to point synchronous optical network (SONET) / Synchronous digital hierarchy (SDH) links, wavelength channels in an optical transport network (OTN) or even dark fiber applications.
GFP provides a standard means of mapping, in a very efficient manner a wide variety of data signals into SONET/SDH frames, enabling compliant equipment from different manufactures to transport both traditional and nontraditional data signals over a SONET/SDH infrastructure. Generic Framing Procedure allows the implementation of multiple transport modes that may coexist within the same transport channel. One mode referred as Frame-Mapped GFP, is optimized for packet switching environments where resource management functions are delegated to the native data clients. This is the transport mode used for native Point-to-Point Protocol (PPP), IP, multiple protocol label switching (MPLS), or Ethernet traffic. The second mode referred as Transparent-Mapped GFP, is intended for delay-sensitive storage area networks (SAN) applications that require bandwidth efficiency and transparency to the line code data. This is the transport mode used by Fiber Channel, FICON and ESCON traffic. The current GFP specification is a result of a joint standardization effort in both American National Standard Institute (ANSI) and International Telecommunication Union ? Telecommunication Standardization Union (ITU-T).