Aug 2026· Asia-Pacific Workshop on Networking· 0 citations· 29 references
Computer Science
Abstract
Modern CDN vendors typically provide multi-server services for content deliveries, in which different requests from the same client are often scheduled to diverse CDN nodes. In this case, the requested data (e.g., short video chunks) always suffer from cold start during its transmission, especially for the first flow of established connections, due to the unknown client-side network quality. Existing schemes focus on single-server startup parameter optimization, whose referred metrics appear to be extremely limited. To address this issue, we propose hStart, a novel collective service paradigm that enables hot-start congestion control by efficient collaboration between CDN nodes. hStart synchronizes probed network quality information (e.g., bandwidth, latency) across CDN nodes to initialize congestion control from historical data, avoiding probing convergence. It uses hierarchical state synchronization to limit overhead, and adaptive prediction with a time-decay trust model to handle dynamics. We implement and deploy the hStart system for the QUIC protocol. Extensive evaluations in both controlled testbeds and large-scale production CDN environments demonstrate that hStart effectively eliminates cold start penalties, boosting 46.47% throughput gain and 21.43% rebuffering reduction.
Spectrum admission in low-latency mobile networks increasingly relies on fast control loops under time-varying traffic and radio conditions. A key challenge in such settings is reconfiguration churn: small score fluctuations near the admission boundary can repeatedly flip the marginal admitted request, even when the resulting utility gain is negligible. These boundary-level replacements are disproportionately expensive because admission and eviction trigger higher-layer control procedures, signaling exchanges, and coordination overhead. To address this problem, we propose TOA-S, a churn-aware admission primitive for latency-bounded mobile control. TOA-S targets cellular radio access network (RAN) deployments operating over licensed spectrum, where control loops of the ultra-reliable low-latency communication class at the near-real-time RAN Intelligent Controller must complete within sub-second decision windows. TOA-S preserves the greedy allocation structure for the responsive core of the admitted set and applies stabilization only at the admission boundary, in a single pass per decision epoch, without iterative optimization or learning. We show that TOA-S incurs only an $\varepsilon $ -bounded utility deviation, modifies at most one membership decision per epoch, and suppresses oscillatory replacements under persistent boundary near-tie conditions. Simulations on synthetic workloads with measurement-verified parameters show that TOA-S substantially reduces reconfiguration churn while maintaining spectrum utilization and latency-compatible execution.
Chi-Jen Wu· IEEE Open Journal of the Com...· 0 citations
This work introduces Online Pricing-based Slice Admission Control and Resource Allocation (OPA) framework, which dynamically assigns pseudo-prices to resources that capture long-term scarcity and anticipated inter-temporal opportunity costs and designs an exponential pricing strategy that guarantees bounded worst-case performance.
Muhammad Sulaiman, Bo Sun, M. A. Salahuddin et al.· 0 citations
SprayCast is proposed, a congestion-adaptive native RDMA multicast scheme for dynamic sparse token Dispatch that uses in-band network telemetry feedback to steer replication away from congested multicast branches and range-based negative acknowledgments for localized loss recovery, saving bandwidth and reducing tail latency in dynamic all-to-all communication.
Yingying Zeng, Ning Xu, Xiaobin Tan et al.· Asia-Pacific Workshop on Net...· 0 citations
The development of immersive video service and large-scale cluster computing technology further expand the potential application scope of time-sensitive networks (TSN). In the delivery network for these emerging services, Ultra-Service Flows (USFs), characterized by ultra-high bandwidth and deterministic latency, have become the most representative traffic type. Therefore, the route scheduling for hybrid deployment of Regular-Service Flows (RSFs) and USF has become an unavoidable issue within a deterministic domain. However, existing research has not thoroughly investigated routing issues for the hybrid deployment of USF and RSF since the significant differences between them. To resolve this issue, a multi-objective optimization model is designed in this paper, in which three key factors are comprehensively considered: the path blocking degree of USF, the available bandwidth rate, and the end-to-end latency of RSF. Subsequently, we propose a cooperative framework where a Transformer-DRL agent, enforced by validity-constraint masking, generates high-quality initial populations to “warm start” NSGA-II. This hybrid design replaces random initialization, effectively resolving the evolutionary “cold start” issue in large-scale topologies while ensuring routing feasibility. The simulation results demonstrate that the algorithm proposed here in significantly improves performance and generalization capabilities, improving the RSF’s overall latency, the USF’s path-blocking degree, and the available bandwidth rate by 10.526%, 14.102%, and 14.286%, respectively.
Mengjie Guo, Qiang Wu, Ran Wang et al.· IEEE Transactions on Network...· 0 citations
The proposed Multi-Path Multi-Level Feedback Queueing (MP-MLFQ) leverages the spatial diversity and regularity of DCNs to realize a scheduler with numerous logical priority levels while occupying as low as 2 physical priority queues within network switches.
Alessandro Cornacchia, Andrea Bianco, Paolo Giaccone et al.· 0 citations
To achieve high-throughput and low-latency Remote Direct Memory Access (RDMA) communication in data center networks, load balancing is critical for preventing congestion and ensuring that traffic is efficiently distributed across available network paths. However, existing schemes may not effectively detect rerouting opportunities in continuous RDMA packet streams and may degrade in-order delivery, limiting their applicability to RDMA traffic. To address these limitations, we propose ProLet, a load balancing scheme that enables proactive probing and reroutes elephant flows at flowlet granularity in lossless RDMA networks. ProLet dynamically fine-tunes per-destination top-of-rack timeouts and enables effective in-network flowlet identification based on real-time network conditions. Meanwhile, it leverages lightweight mice flows as proactive probes to maintain network-wide congestion awareness. This allows ProLet to reroute elephant flows before congestion accumulates, mitigating the persistent queue buildup inherent in subflow-based schemes. Extensive numerical evaluations demonstrate that ProLet reduces average and tail flow completion time slowdowns by 69% and 79%, respectively, compared to state-of-the-art load balancing schemes.
Hong Wang, Jinhao Luo, J. Tan et al.· Asia-Pacific Workshop on Net...· 0 citations