Name
Abstract | ||
|---|---|---|
Maximizing bandwidth utilization of optical onchip interconnects is essential to compensate for static power overheads in optical networks-on-chip. Shared optical buses were shown to be a power-efficient, modular design solution with tremendous power saving potential by allowing optical bandwidth to be shared by all connected nodes. Previous proposals resolve bus contention by scheduling senders sequentially on the entire optical bandwidth; however, logically splitting a bus into subchannels to allow both sequential and parallel data transmission has been shown to be highly efficient in electrical interconnects and could also be applied to shared optical buses. In this paper, we propose an efficient subchannel scheduling algorithm that aims to minimize the number of bus utilization cycles by assigning sender-receiver pairs both to subchannels and time slots. We present both a distributed and a centralized bus arbitration scheme and show that both can be implemented with low overheads. Our results show that subchannel scheduling can more than double throughput on shared optical buses compared to sequential scheduling without any power overheads in most cases. Arbitration latency overheads compared to state-of-the-art sequential schemes are moderate-to-low for significant bus bandwidths and only noticeable for low injection rates. |
Year | DOI | Venue |
|---|---|---|
2017 | 10.1109/HOTI.2017.18 | 2017 IEEE 25th Annual Symposium on High-Performance Interconnects (HOTI) |
Keywords | Field | DocType |
optical onchip interconnects,static power overheads,optical networks-on-chip,modular design solution,tremendous power saving potential,entire optical bandwidth,bus utilization cycles,sequential scheduling,Shared Optical On-chip Buses,bus contention,bandwidth utilization maximization,subchannel scheduling algorithm,sequential data transmission,parallel data transmission | Data transmission,Latency (engineering),Scheduling (computing),Waveguide (optics),Computer science,Parallel computing,Computer network,Bandwidth (signal processing),Bus contention,Throughput,Modular design | Conference |
ISBN | Citations | PageRank |
978-1-5386-1014-5 | 0 | 0.34 |
References | Authors | |
20 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Sebastian Werner | 1 | 46 | 8.92 |
| Javier Navaridas | 2 | 201 | 23.58 |
| Mikel Luján | 3 | 167 | 20.01 |