Shortest propagation delay (SPD) first scheduling for EPONs with heterogeneous propagation delays

Abstract

Abstract-Due to the geographic distribution of its subscribers, Ethernet Passive Optical Networks (EPONs) have typically varying propagation delays between the Optical Network Units (ONUs) and the Optical Line Terminal (OLT). In this paper, we consider EPONs with an offline scheduling framework, which enables Quality-of-Service mechanisms by collecting bandwidth requests from all ONUs before the OLT makes dynamic bandwidth allocations for transmissions on the shared ONUs-to-OLT upstream channel. We propose and evaluate the Shortest Propagation Delay (SPD) first scheduling policy which sequences the ONUs' upstream transmissions in increasing order of the ONUs' propagation delays, i.e., the upstream transmission of the ONU with the smallest propagation delay is scheduled first. We formally analyze the competitiveness of SPD first scheduling and find that it achieves very close to optimal performance. We characterize the stability limit for Gated and Limited grant sizing in conjunction with SPD grant scheduling. We evaluate the cycle length and packet delay with SPD scheduling through probabilistic analysis and simulations and find significant reductions in packet delay with SPD first scheduling in EPONs with heterogeneous propagation delays, especially when Limited grant sizing is employed.

Offline and Online Multi-Thread Polling in Long-Reach PONs: A Critical Evaluation

“…Grant scheduling has been intensively researched for increasing efficiency by avoiding voids (idle times) between ONU upstream transmissions [47]- [50] and providing quality-of-service differentiation [51], [52]. We employ the scheduling policy from [50] to mitigate idle times due to different round-trip propagation delays.…”

Investigation of the DBA Algorithm Design Space for EPONs

“…The following are some grant scheduling policies that have been proposed and evaluated in the literature. 1) Shortest grant or shortest processing time (SPT) first [37] orders the ONUs in ascending order by grant size 2) Largest number of frames (LNF) first [11] orders the ONUs in descending order by number of queued frames 3) Shortest propagation delay (SPD) first [32] orders the ONUs in ascending order by round-trip propagation delay.…”

“…We simulated an EPON with a channel capacity, Gb/s and ONUs. We varied the maximum propagation delay to represent three different EPON reaches: 1 to 10 km (6.67 to 50 s), 1 to 50 km (6.67 to 250 s), and 1 to 100 km (6.67 to 500 s) (in [32], we illustrate the feasibility of these ranges in practical EPON architectures). A quad modal packet size distribution was used for all simulation experiments: 60% 64 bytes, 4% 300 bytes, 11% 580 bytes, and 25% 1518 bytes.…”

“…For instance, a few studies [9]- [13] have focused on the overall scheduling framework that triggers bandwidth allocation decisions. Many studies have focused on sizing the bandwidth grants for the individual ONUs [1], [14]- [22], while others have primarily focused on scheduling the sized grants [23]- [32].…”

“…Importantly, some of these novel DBA algorithms outperform all previously proposed DBA algorithms. For instance, combining the double-phase polling (DPP) scheduling framework [9], [10] with limited with excess distribution grant sizing [14], [33] and shortest propagation delay (SPD) first scheduling [32] achieves better throughput delay performance than previously examined DBA algorithms. Further building on this (DPP, Limited with excess, SPD) DBA algorithm, we propose a novel excess bandwidth sharing strategy that further significantly improves throughput-delay performance.…”