Methods and apparatuses for processing data channels
A system and method for processing an extremely high data rate datastream. Embodiments of the invention provide methods for performing various operations on the datastream in order to map the datastream from one protocol to another as well as providing methods for processing multiple channels of a given protocol. For one embodiment of the invention a portion of a datastream is received to a stream buffer. A data stream window is then created from the received portion, the data stream window containing data of more than one protocol. A corresponding portion of the data of a same protocol, is accessed through each of a plurality of processing machines. The accessed data is concurrently processed at each of the processing machines.
1. A system for processing data comprising:
an integrated circuit including:
means for receiving and transmitting one or more datastreams comprising data to be processed;
a memory for buffering the one or more datastreams the buffering effected at an atomic level as small as one bit at a physical wire level of the integrated circuit;
a scheduler for scheduling processing of the data upon configured criteria being met; and
a plurality of processing machines; and means for shifting a datastream through a datastream window, the means for shifting comprising an integrated circuit hardware shifting mechanism, the datastream window having multiple access points that allow the plurality of processing machines to access corresponding portions of the datastream and effect concurrent processing of the respective portions of the datastream.
2. The system of claim 1 wherein the access points are programmed to access any portion within the datastream window.
3. The system of claim 1 , wherein the processing is selected from the group consisting of synchronization, analysis, field extraction, data manipulation, output stream construction and combinations thereof.
4. The system of claim 1 wherein multiple datastreams are processed by common shared logic, a datastream context fetched from a storage mechanism during a processing period and stored to the storage mechanism upon completion of the processing period.
5. The system of claim 4 wherein the processing is selected from the group consisting of synchronization, analysis, field extraction, data manipulation, output stream construction and combinations thereof.
6. The system of claim 1 wherein the means for receiving one or more datastreams comprises a communication processing engine and the means for shifting a datastream through a datastream window comprises a shift register.
7. A system comprising:
a plurality of concatenated data processing systems, each data processing system implemented within an integrated circuit, and including:
means for receiving and transmitting one or more datastreams;
means for buffering the one or more datastreams, the buffering effected at an atomic level as small as one bit at a physical wire level of the integrated circuit;
means for scheduling processing of the data upon configured criteria being met; and
means for shifting a datastream through a datastream window, the means for shifting comprising an integrated circuit hardware shifting mechanism, the datastream window containing data of multiple protocols, the datastream window having multiple access points that allow a plurality of processing machines to access corresponding portions of the datastream, each of the plurality of processing machines concurrently processing the corresponding portion of the datastream.
8. The system of claim 7 wherein multiple datastreams are processed by common shared logic, a datastream context fetched from a storage mechanism during a processing period and stored to the storage mechanism upon completion of the processing period.
9. The system of claim 8 wherein the processing is selected from the group consisting of synchronization, analysis, field extraction, data manipulation, output stream construction and combinations thereof.
10. A method comprising:
receiving a portion of a datastream to a stream buffer, the portion of the datastream containing data associated with a first channel of a plurality of channels, the data buffered at an atomic level as small as one bit at an integrated circuit physical wire level;
retrieving the datastream context of the first channel associated with the received portion of the datastream; and
upon configured criteria being met, processing data from the first channel using the retrieved context, the processing effected using one of a plurality of processing machines; and
concurrently processing data from one or more other channels of the plurality of channels, wherein the stream buffer, plurality of channels, and the plurality of processing machines are implemented within an integrated circuit.
11. The method of claim 10 wherein the context is switched upon receiving a subsequent portion of the data stream.
12. The method of claim 11 wherein the data to be processed comprises a specified number of events accumulated in a channel buffer.
13. The method of claim 12 wherein the specified number of events is based upon memory resources.
14. The method of claim 10 wherein each of the plurality of channels implements a protocol of a plurality of protocols.
15. The method of claim 14 wherein the plurality of protocols include protocols selected from the group consisting of T1, E1, J1, T3, E3, SONET, SDH, Ethernet, GFP, HDLC, PPP, Frame Relay, ATM, MLPPP, MLFR, MPLS, PWE3, IP, POS, EOS, Fiber Channel, Token Ring, FDDI, SCSI, SAS, iSCSI, TCP, UDP, FICON, and DSL.