EP-4742544-A1 - CONDITIONALLY PROCESSING COMPUTE RESOURCES

Abstract

Systems and techniques for conditionally processing compute resources are described herein. A data payload is received. Parity data is removed from the data payload to generate a codeword dataset. Frames are identified in the codeword dataset. Headers of the frames are evaluated using a conditional decoding map. The frames are marked based on a destination in the headers found in the conditional decoding map. A set of target codewords are identified from the frames based on the marking. Remaining frames of the frames outside the set of target codewords are discarded. The set of target codewords is transmitted to a buffer of a conditional decoder. Codewords from the buffer are decoded using the conditional decoder.

Inventors

• BERNARD, Christopher, Thomas
• NOTCH, Scott

Assignees

• CALIX, INC.

Dates

Publication Date

20260513

Application Date

20251103

Claims (15)

1. A system for conditionally processing compute resources, comprising: at least one processor; and memory comprising instructions that, when executed by the at least one processor, causes the at least one processor to perform operations to: receive a data payload; remove parity data from the data payload to generate a codeword dataset; identify frames in the codeword dataset; evaluate headers of the frames using a conditional decoding map; mark the frames based on a destination in the headers found in the conditional decoding map; identify a set of target codewords from the frames based on the marking; discard remaining frames of the frames outside the set of target codewords; transmit the set of target codewords to a buffer of a conditional decoder; and decode codewords from the buffer using the conditional decoder.
2. The system of claim 1, wherein the buffer absorbs packet-delay variance from a high-speed physical layer rate to a reduced rate of a network slice.
3. The system of claim 1, the memory further comprising instructions that, when executed by the at least one processor, causes the at least one processor to perform operations to: aggregate the frames prior to evaluation of the headers, wherein the aggregation uses an inter-frame gap.
4. The system of claim 1, the memory further comprising instructions that, when executed by the at least one processor, causes the at least one processor to perform operations to: encapsulate the frames into a single XGPON Encapsulation Method, XGEM, frame prior to generating the codeword dataset.
5. The system of claim 1, the memory further comprising instructions that, when executed by the at least one processor, causes the at least one processor to perform operations to: size a compute resource of the optical network terminal according to throughput of a network slice of the optical line terminal.
6. The system of claim 5, wherein the instructions to size the compute resource comprises instructions to: track the slice throughput; and dynamically adjust a capacity of the compute resource based on the slice throughput.
7. The system of claim 1, the memory further comprising instructions that, when executed by the at least one processor, causes the at least one processor to perform operations to: increase capacity for a compute resource of the optical line terminal based on a bit-error-rate of the conditional decoder.
8. A method for conditionally processing compute resources, comprising: receiving a data payload; removing parity data from the data payload to generate a codeword dataset; identifying frames in the codeword dataset; evaluating headers of the frames using a conditional decoding map; marking the frames based on a destination in the headers found in the conditional decoding map; identifying a set of target codewords from the frames based on the marking; discarding remaining frames of the frames outside the set of target codewords; transmitting the set of target codewords to a buffer of a conditional decoder; and decoding codewords from the buffer using the conditional decoder.
9. The method of claim 8, wherein the buffer absorbs packet-delay variance from a high-speed physical layer rate to a reduced rate of a network slice.
10. The method of claim 8, further comprising: aggregating the frames prior to evaluating the headers, wherein the aggregation uses an inter-frame gap.
11. The method of claim 8, further comprising: encapsulating the frames into a single XGPON Encapsulation Method, XGEM, frame prior to generating the codeword dataset.
12. The method of claim 8, further comprising: sizing a compute resource of the optical network terminal according to throughput of a network slice of the optical line terminal.
13. The method of claim 12, wherein sizing the compute resource comprises: tracking the slice throughput; and dynamically adjusting a capacity of the compute resource based on the slice throughput.
14. The method of claim 8, further comprising: increasing capacity for a compute resource of the optical line terminal based on a bit-error-rate of the conditional decoder.
15. At least one machine-readable medium comprising instructions for conditionally processing compute resources that, when executed by at least one processor, cause the at least one processor to perform the method of any one of claims 8 to 14.

Description

This application claims the benefit of priority to U.S. Provisional Patent Application Serial No. 63/716,802, filed on November 6, 2024. TECHNICAL FIELD Embodiments described herein generally relate to network processing power consumption reduction and, in some embodiments, more specifically to forward error correction power conservation. BACKGROUND Low density parity check forward error correction (LDPC-FEC) can be used to determine network data error rates. LDPC-FEC can encode data resulting in processing power (e.g., electric energy) usage to encode and decode data. In some instances, decoding can be responsible for more than half of the processing workload and power consumption of a networking device. Network standards bodies have sought to improve the energy impact of operating access networks. Worldwide energy crises have a renewed the call to implement reduced energy consumption in networking hardware executing as central office (CO) equipment and as customer premise equipment (CPE). SUMMARY The invention is a system, method and machine-readable medium as defined in the appended claims. BRIEF DESCRIPTION OF THE DRAWINGS In the drawings, which are not necessarily drawn to scale, like numerals can describe similar components in different views. Like numerals having different letter suffixes can represent different instances of similar components. The drawings illustrate generally, by way of example, but not by way of limitation, various embodiments discussed in the present document. FIG. 1 illustrates an example of a relation between power consumption and throughput for typical optical network unit (ONU) and a relation between power consumption and throughput for an ideal ONU.FIG. 2 is a block diagram of an example of an environment and a system for conditional codeword decoding, according to an embodiment.FIG. 3 illustrates a flow diagram of an example of downstream service data unit (SDU) mapping into physical layer (PHY) frames.FIG. 4 is a block diagram of an example of an environment and a system for conditionally processing compute resources, according to an embodiment.FIG. 5 illustrates an example of a process for ten gigabit passive optical network (XGPON) Encapsulation Method (XGEM) look-ahead for conditionally processing compute resources, according to an embodiment.FIG. 6 illustrates and example of a framing sublayer (FS) packet for conditionally processing compute resources, according to an embodiment.FIG. 7 illustrates and example of a ten gigabit passive optical network (XGPON) Encapsulation Method (XGEM) packet for conditionally processing compute resources, according to an embodiment.FIG. 8 illustrates an example of a reduced operations process for ten gigabit passive optical network (XGPON) Encapsulation Method (XGEM) look-ahead for conditionally processing compute resources, according to an embodiment.FIG. 9 is a flow diagram of an example method for conditionally processing compute resources, according to an embodiment.FIG. 10 is a block diagram of an example network interface device for conditionally processing compute resources, according to an embodiment.FIG. 11 is a block diagram of an application specific integrated circuit (ASIC) for conditionally processing compute resources, according to an embodiment.FIG. 12 is a block diagram of frame reconstruction of an example frame for conditionally processing compute resources, according to an embodiment.FIG. 13 is a block diagram of that illustrates decoding operations for various packet configurations for conditionally processing compute resources, according to an embodiment.FIG. 14 is a flow diagram of an example method for conditionally processing compute resources, according to an embodiment.FIG. 15 is a block diagram illustrating an example of a machine upon which one or more embodiments can be implemented. DETAILED DESCRIPTION FIG. 1 illustrates an example of a relation between power consumption and throughput for typical optical network unit (ONU) 105 and a relation between power consumption and throughput for an ideal ONU 110. As shown in FIG. 1, for the typical ONU 105, there is little power savings achieved when traffic stops flowing. The ideal ONU 110 illustrates power going to a minimal amount when there is no traffic. The systems and techniques described herein measurably increase the power consumption to throughput slope of the typical ONU 105 to approach the ideal ONU 110. The systems and techniques described herein make modification to a passive optical network transmission convergence (PON-TC) layer to achieve better power/traffic correlation. International Telecommunication Union (ITU) and Institute of Electrical and Electronics Engineers (IEEE) PON-TC layers beyond physical layer (PHY) rates of 10 gigabits (Gbs) use more advanced forward error correction (FEC) coding gain to meet industry accepted optical distribution network (ODN) link budgets. In the ITU (G.9804.2) High-Speed PON standard, low density parity check