strict-priority delay modeling

This work addresses a critical limitation in existing latency models for strict-priority scheduling, which neglect the impact of the transmit ring buffer (TXR) at switch egress ports, leading to inaccurate delay estimates for high-priority packets. For the first time, this study incorporates TXR into the end-to-end delay analysis framework by proposing a method to measure TXR size and integrating deterministic and stochastic modeling, network measurements, and hardware behavior characterization. The resulting model significantly improves the accuracy of both worst-case delay bounds and delay distribution predictions for high-priority traffic. Validated across multiple commercial off-the-shelf switches, the approach effectively bridges the gap between theoretical models and real hardware behavior, thereby enhancing the reliability of real-time network system design.

This work addresses the well-known conservatism of packet delay bounds derived from virtual delay in classical network calculus, which often fail to accurately capture the actual worst-case delay. Revisiting packet delay analysis, the paper proposes a novel upper bound that relies solely on arrival and service curves without requiring additional assumptions. The authors theoretically prove that the maximum packet delay is always no greater than the maximum virtual delay and leverage this insight to derive a tighter delay bound that strictly improves upon classical results. Experimental evaluation in time-sensitive networking (TSN) scenarios demonstrates that the proposed bound significantly enhances analytical accuracy.

Modeling task success probability in computation-centric networks is challenging due to dynamic task arrivals, server capacity constraints, and coupling between bidirectional link delays. Method: We propose the first unified analytical framework: (i) deriving the first closed-form expression for task success probability; (ii) establishing tight upper and lower bounds; (iii) jointly modeling Erlang blocking, stale state information, and stochastic end-to-end delay; (iv) characterizing network and queueing delays via Laplace transforms; and (v) abstracting forwarding policies as replaceable probabilistic parameters—enabling compatibility with diverse policies and delay distributions. Contribution/Results: The framework is broadly applicable and highly scalable. Across extensive parameter regimes, theoretical predictions match simulations with exceptional accuracy: bound errors are ≤0.01 (lower) and ≤0.016 (upper). This significantly improves prediction fidelity for task success probability and enhances interpretability in system design.

This work addresses the challenge of guaranteeing hard per-packet end-to-end deadlines and throughput in multi-hop wireless networks under interference. The authors propose a network slicing–based modeling approach that decouples inter-flow queue dynamics and demonstrates that queue stability alone is insufficient to ensure strict deadline compliance. By establishing a precise relationship between end-to-end delay and link scheduling intervals, they introduce the generalized pinwheel scheduling problem into multi-hop wireless networking for the first time. Building on this insight, they design a decentralized, polynomial-time scheduling algorithm that provably meets hard deadlines under arbitrary interference models while achieving near-optimal throughput performance.

To address the challenge of deterministic low-latency scheduling for periodic message transmission between antennas and remote processing units in Cloud-RAN—where strict protocol deadlines must be met while avoiding buffering and collision delays induced by statistical multiplexing—this paper pioneers the application of deterministic conflict-free scheduling to Cloud-RAN fronthaul networks. We propose two algorithms: (i) an analytical zero-buffer scheduling algorithm tailored for short-path or light-load scenarios, and (ii) PMLS (Periodic Message Latency Scheduling), a heuristic algorithm supporting bounded buffering. We theoretically prove that a zero-buffer feasible schedule always exists under short-path or low-load conditions. Experimental results demonstrate that PMLS achieves zero-delay deterministic schedules with high probability even under full load, significantly enhancing latency predictability and resource utilization.

This study addresses the reliance of delay computation in Time-Sensitive Networking (TSN) on global knowledge and the poor interoperability among heterogeneous shapers. To overcome these limitations, this work proposes a decentralized delay analysis framework based on local information. By formally defining "delay segments," the framework enables standardized interactions across multiple shapers and provides local pseudocode implementations for algorithms such as Strict Priority and Credit-Based Shaping. Experimental evaluations assess five models in terms of delay, jitter, and flow admission efficiency, validating the suitability of different shapers for specific network topologies and traffic distributions. Ultimately, this research offers an efficient and scalable analytical methodology for heterogeneous scheduling in TSNs.

Existing models for 5G HARQ delay violation probability (DVP) significantly underestimate the true DVP under stringent latency constraints, as they neglect critical delay components—including queuing, transmission, decoding, feedback, and control signaling—and fail to account for the practical HARQ behavior that allows new packets to be transmitted in parallel without waiting for ACKs. This work presents the first DVP model that fully incorporates the timing characteristics and parallel transmission mechanisms of 5G HARQ. By integrating queueing theory with Markovian analysis and adhering to 3GPP-specified timing assumptions, we derive a tight upper bound on DVP. Extensive ns-3 simulations using the 5G-LENA framework demonstrate that this bound substantially outperforms existing simplified models in low-latency scenarios, accurately capturing the system’s real-world performance.

This work addresses load balancing in parallel infinite-server queues under action delays by explicitly incorporating delay into the state representation for the first time. The authors model the delay process using an Erlang phase-type structure, thereby constructing a finite-dimensional Markov jump system and employing ordinary differential equations to explicitly track tasks in transit. Exploiting symmetry between two servers, the system dynamics are reduced to a single mode capturing load imbalance. Theoretical analysis establishes equivalence between this formulation and the delayed-information model in their linearized dynamics, overcoming limitations of traditional delay-differential-equation approaches. The study derives the characteristic equation governing imbalance dynamics for an arbitrary number of phases, and numerical experiments confirm the accuracy of the fluid approximation while quantifying the effects of phase count, routing sensitivity, and mean delay on transient response.

本文针对通信网络中由指数分布延迟触发的数据包流问题，引入M^T/G/1排队模型，并利用矩阵解析方法求解相关性能指标。

研究了具有潜在Erlang因子依赖子任务时间的两服务器拆分-合并队列问题，通过精确分析得到完成时间和平均等待时间，并证明独立性会高估平均等待时间。