Scientific workflows are fundamental for automating complex data processing tasks. Research on workflow scheduling and execution strategies often relies on real-world workflows for testing and validation. The setup of such workflows is timeconsuming and error-prone due to their dependence on large datasets, specialized software, or specific hardware requirements. Synthetic workflows have been proposed as a solution, offering realistic benchmarking scenarios. Existing workload generators, however, primarily model tasks based on aggregated resource consumption metrics, such as total CPU time or peak memory usage, which do not capture the dynamic resource usage patterns exhibited by real-world tasks. This limits the expressiveness of experiments performed on synthetic workflows. In this work, we introduce a novel approach for generating synthetic workflow tasks that capture dynamic resource usage patterns. In addition, we derive realistic task models by monitoring real workflow task executions and applying segmented linear regression to the measurement data. These models serve as the basis for generating synthetic tasks. Our method allows researchers to create synthetic workflows that more accurately reflect the behavior of realworld scientific workflows, while still being efficient and without requiring external datasets.

Efficient hardware-supported packet scheduling is a cornerstone of fine-grained packet prioritization in modern networking systems. Algorithms such as Weighted Fair Queuing and Shortest Job First offer theoretical guarantees for achieving flow fairness and minimal flow completion times; however, their underlying priority-queue operation is difficult to implement at line rate due to its complexity. In this paper, we propose Dancing-Q, a line-rate dynamic approximation of a priority queue that is supported by current hardware and provides optimality guarantees. Essentially, Dancing-Q mimics the packet-sorting operation of a priority queue by distributing packets according to their ranks across multiple dynamically assigned FIFO queues. Key to the optimality guarantee are dynamic queue bounds that map ranges of packet ranks to the FIFO queues which are served using a strict priority scheduler. To minimize rank inversions, i.e., out-of-order scheduling when compared with an ideal priority queue, we introduce a per-queue local optimization strategy that rapidly drives queue bounds to optimal values based on the empirically observed packet ranks. Owing to the per-queue locality and a low memory and computation overhead, this optimization strategy allows prototyping Dancing-Q on programmable switches and SmartNICs. We demonstrate the priority queue emulation of Dancing-Q on a Tofino-based ASIC and use large-scale ns-3 experiments to show that compared to standard methods such as SP-PIFO, Dancing-Q (i) attains the optimal rank inversions, (ii) mimics the queuing delays of a priority queue on a significantly larger rank range, and (iii) curbs large flow starvation. Empirical evaluations show that Dancing-Q minimizes the flow completion time within a 4.4% overhead of the ideal priority queue.

© 2026 IFIPThis demonstration showcases how integrating the IETF Deterministic Networking (DetNet) Packet Replication, Elimination, and Ordering Functions (PREOF) into the 6G architecture improves resilience and latency in an industrial scenario. Our setup connects two synchronized robotic arms via two disjoint and heterogeneous 6G networks, one employing a hardware-based and the other a software-based User Plane Function (UPF). Packets are replicated across both paths and eliminated upon reception, ensuring robust communication even in the presence of link or component failures. Additionally, latency is minimized as the receiver forwards the first arriving packet.

Modern scientific data analysis workflows frequently operate across diverse, distributed infrastructure. In such environments, task placement and data transfers between them can impact overall performance significantly. While typical scheduling approaches often focus on computational resource allocation, the impact of network constrained data transfers is often ignored or simplified. In this work-in-progress paper, we propose an approach that explicitly models data transfers while considering the network topology. Integrating topology information with data size estimates and node capability insights, we develop a cost function that estimates the makespan. It considers all previously mentioned factors, for a specified task and data placement. We formulate the task of finding a good assignment as a minimization problem and outline a genetic-algorithm-based search strategy to approximate the best placement. The proposed framework aims to enable topology-aware scheduling decisions, or to refine schedules generated elsewhere, for distributed scientific workflows.

The use of redundant data transmission for critical applications is gaining increasing traction because it increases reliability and timing stability. However, it also allows novel methods to detect anomalies specific to such redundant transmissions, but these concepts have not been explored so far. Therefore, we present and implement RCAND (Redundant Channel ANomaly Detection), the first approach to detect anomalies in redundant transmissions on multiple channels. We evaluate RCAND in a series of lab experiments, in which we first quantify measurement precision and component delays, and subsequently demonstrate the anomaly detection capabilities of RCAND, including the detection of a machine-in-the-middle (MITM) attack.

While detailed resource usage monitoring is possible on the low-level using proper tools, associating such usage with higher-level abstractions in the application layer that actually cause the resource usage in the first place presents a number of challenges. Suppose a large-scale scientific data analysis workflow is run using a distributed execution environment such as a compute cluster or cloud environment and we want to analyze the I/O behaviour of it to find and alleviate potential bottlenecks. Different tasks of the workflow can be assigned to arbitrary compute nodes and may even share the same compute nodes. Thus, locally observed resource usage is not directly associated with the individual workflow tasks. By acquiring resource usage profiles of the involved nodes, we seek to correlate the trace data to the workflow and its individual tasks. To accomplish that, we select the proper set of metadata associated with low-level traces that let us associate them with higher-level task information obtained from log files of the workflow execution as well as the job management using a task orchestrator such as Kubernetes with its container management. Ensuring a proper information chain allows the classification of observed I/O on a logical task level and may reveal the most costly or inefficient tasks of a scientific workflow that are most promising for optimization.

The Internet is a dynamic network of networks utilizing various protocols across different layers to enable end-to-end communication between processes. Packet loss can occur for numerous reasons at any hop during a packet's journey. This paper explores the influence of packet size on the likelihood of packet loss. Discovering such a relationship provides valuable insights into optimal packet sizing, ultimately influencing the design of future protocols, applications, and networks. This study identifies correlations between packet size and packet loss through probe packets of varying sizes by leveraging public Internet services like ICMP and DNS for data collection. Our key contributions include a large-scale experiment gathering packet loss data from tens of thousands of targets, a dataset of packet loss probabilities across different packet sizes, and a detailed dataset analysis. The analysis shows that while packet size has a statistically significant influence on packet loss, its overall predictive strength is relatively low.

In the future, Low-Earth Orbit satellite constellations will directly provide cellular services from space. Especially for large-scale disasters, where desperately needed communication infrastructure is often severely damaged or destroyed, this may be a significant game-changer for disaster relief efforts. However, limited or partial availability of satellite connectivity may be more probable than full coverage in the next years. This paper proposes and investigates a possible integration of direct-to-cell satellite communication in decentralized disaster networks, facilitating a combination of delay-tolerant networking with limited satellite availability. Simulation results indicate that already a few satellite links positively impact communication performance, increasing network connectivity and message spread in a city-wide disaster scenario with user mobility.

Segmented regression is a statistical method that approximates a function f by a piecewise function (f) over cap using noisy data samples. Min-epsilon approaches aim to reduce the regression function's mean squared error (MSE) for a given number of k segments. An optimal solution for min-. segmented regression is found in O(n(2)) time (Bai & Perron, 1998; Yamamoto & Perron, 2013) for n samples. For large datasets, current heuristics improve time complexity to O(n log n) (Acharya et al., 2016) but can result in large errors, especially when exactly k segments are used. We present a method for min-epsilon segmented regression that combines the scalability of top existing heuristic solutions with a statistical efficiency similar to the optimal solution. This is achieved by using a new method to merge an initial set of segments using precomputed matrices from samples, allowing both merging and error calculation in constant time. Our approach, using the same samples and parameter k, produces segments with up to 1,000x lower MSE compared to Acharya et al. (2016) in about 100x less runtime on datasets over 10(4) samples.

Microblogging is a crucial mode of online communication. However, launching a new microblogging platform remains challenging, largely due to network effects. This has resulted in entrenched (and undesirable) dominance by established players, such as X/Twitter. To overcome these network effects, Bluesky, an emerging microblogging platform, introduced starter packs — curated lists of accounts that users can follow with a single click. We ask if starter packs have the potential to tackle the critical problem of social bootstrapping in new online social networks. We assess whether starter packs have indeed been helpful in supporting Bluesky growth. Our dataset includes 25.05 × 10⁶ users and 335.42 × 10³ starter packs with 1.73 × 10⁶ members, covering the entire lifecycle of Bluesky. We study the usage of these starter packs, their ability to drive network and activity growth, and their potential downsides. We also quantify the benefits of starter packs for members and creators on user visibility and activity while identifying potential challenges. By evaluating starter packs’ effectiveness and limitations, we contribute to the broader discourse on platform growth strategies and competitive innovation in the social media landscape.

The growing demands of ultra-reliable and low-latency communication (URLLC) in 5G networks necessitate enhanced resilience mechanisms to address user plane failures caused by outages, hardware defects, or software bugs. An important aspect for achieving ultra-reliable communication is the redundant transmission of packets, as also highlighted in 3GPP Release 18. This paper explores leveraging the Packet Replication, Elimination, and Ordering Function (PREOF) to achieve 1+1 path protection within 5G environments. By extending existing 5G components with mechanisms for path redundancy and offloading the reordering mechanism to external servers, the proposed approach ensures ultra-low latency and minimal packet loss in case of a failure. A conceptual integration of redundant paths and programmable elements is presented, with considerations for deployment in existing 5G infrastructures and the trade-offs of latency versus enhanced traffic engineering. Future work aims to evaluate practical implementations using P4-based hardware and offloading technologies like DPDK and eBPF.

In the recent years, scientific workflows gained more and more popularity. In scientific workflows, tasks are typically treated as black boxes. Dealing with their complex interrelations to identify optimization potentials and bottlenecks is therefore inherently hard. The progress of a scientific workflow depends on several factors, including the available input data, the available computational power, and the I/O and network bandwidth. Here, we tackle the problem of predicting the workflow progress with very low overhead. To this end, we look at suitable formalizations for the key parameters and their interactions which are sufficiently flexible to describe the input data consumption, the computational effort and the output production of the workflow's tasks. At the same time they allow for computationally simple and fast performance predictions, including a bottleneck analysis over the workflow runtime. A piecewise-defined bottleneck function is derived from the discrete intersections of the task models' limiting functions. This allows to estimate potential performance gains from overcoming the bottlenecks and can be used as a basis for optimized resource allocation and workflow execution.

Small satellites used for Earth observation generate vast amounts of high-dimensional data, but their operation in low Earth orbit creates a significant communication bottleneck due to limited contact times and harsh, varying channel conditions. While deep joint source-channel coding (DJSCC) has emerged as a promising technique, its practical application to the complex satellite environment remains an open question. This paper presents a comprehensive DJSCC framework tailored for satellite communications. We first establish a basic system, DJSCC-SAT, and integrate a realistic, multi-state statistical channel model to guide its training and evaluation. To overcome the impracticality of using separate models for every channel condition, we then introduce an adaptable architecture, ADJSCC-SAT, which leverages attention modules to allow a single neural network to adjust to a wide range of channel states with minimal overhead. Through extensive evaluation on Sentinel-2 multi-spectral data, we demonstrate that our adaptable approach achieves performance comparable to using multiple specialized networks while significantly reducing model storage requirements. Furthermore, the adaptable model shows enhanced robustness to channel estimation errors, outperforming the non-adaptable baseline. The proposed framework is a practical and efficient step toward deploying robust, adaptive DJSCC systems for real-world satellite missions.

Time-Sensitive Networking (TSN) enables deterministic and low-latency communication for real-time applications over Ethernet. That is accomplished by leveraging scheduling and shaping techniques configured for each egress port within the network switches. Although Time Aware Shaper (TAS) is a promising solution for TSN, its adoption often involves substantial complexity. In this work, we propose Residence Delay Aggregation (RDA), a novel asynchronous TSN mechanism that offers dynamic traffic scheduling adapted to the traffic load. Specifically, the proposed RDA mechanism provides upper bound delays similar to other asynchronous TSN mechanisms while improving the flexibility of traffic scheduling and reducing the deployment complexity.