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 InterPlanetary File System (IPFS) is the largest decentralized content-centric storage network. While its architecture enables resilient, distributed content delivery, it can be abused to host and disseminate malicious content. Public IPFS HTTP gateways further expand this threat surface, enabling attackers to deploy phishing websites and leverage gateway reputation to evade detection. This model can keep content available even after attackers go offline and challenges traditional phishing detection systems.

Content moderation is a major challenge for online platforms. While user-driven blocking is a common tool, its dynamics are usually hidden as moderation data is private. Bluesky makes moderation actions public-by-design, providing an unprecedented opportunity to study a community-driven moderation ecosystem at scale. We leverage this transparency to (1) map the ecosystem of moderation blocking actions across 34 million users, including both individual blocks and the through blocklists, (2) identify the signals that correlate with blocking, and (3) measure the consequences of these actions. We demonstrate that community blocking is widespread, with a volume several orders of magnitude higher than official takedowns, and affects the visibility of more than 90 % of Bluesky content. The blocked accounts represent the most active, popular, toxic, and politically inclined users. However, different blocklists target different types of accounts and behaviors. Finally, blocking does not decrease the popularity and activity of the blocked users and has a limited effect on the social graph. By quantifying its dynamics and trade-offs, our study provides empirical grounding for designing future moderation systems that are transparent, pluralistic, and resistant to centralized control. Taken together, this study provides the first large-scale, quantitative analysis of a community-driven moderation ecosystem, demonstrating how individual and collective interventions influence user behavior.

Segmented regression is a statistical method that approximates a function $f$ by a piecewise function $\hat{f}$ 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-$\epsilon$ segmented regression is found in $\mathcal{O}(n^2)$ time (Bai & Perron, 1998; Yamamoto & Perron, 2013) for $n$ samples. For large datasets, current heuristics improve time complexity to $\mathcal{O}(n\log{n})$ (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 1,000 times lower MSE compared to Acharya et al. (2016) in about 100 times less runtime on data sets 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 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 recent years, scientific workflows have become increasingly popular. However, their tasks are often seen as black boxes, making it difficult to optimize them or identify bottlenecks due to the complex relationships between tasks. Several factors impact task progress, including input data availability, computing power, data transfer speed and network connectivity. During task execution, resource requirements may change significantly. We propose a new method to model task requirements over their lifetime. Using these models, we predict resource consumption over time and execution duration based on a given allocation strategy with low overhead. This method enables computationally simple and fast performance predictions, including bottleneck analysis during workflow runtime. We derive a piecewise-defined bottleneck function from the discrete intersections of the task models' limiting functions. This allows us to predict potential performance gains when mitigating bottlenecks and aids in better resource allocation and workflow execution.

Analyzing user data while protecting the privacy of individuals remains a big challenge. Trusted execution environments (TEEs) are a possible solution as they protect processes and Virtual Machines (VMs) against malicious hosts. However, TEEs can leak access patterns to code and to the data being processed. Furthermore, when data is stored in a TEE database, the data volume required to answer a query is another unwanted side channel that contains sensitive information. Both types of information leaks, access patterns and volume patterns, allow for database reconstruction attacks.

Scientific workflow management systems enable the reproducible execution of data analysis pipelines on cluster infrastructures managed by resource managers such as Kubernetes, Slurm, or HTCondor. These resource managers require resource estimates for each workflow task to be executed on one of the cluster nodes. However, task resource consumption varies significantly between different tasks and for the same task with different inputs. Furthermore, resource consumption also fluctuates during a task’s execution. As a result, manually configuring static memory allocations is error-prone, often leading users to overestimate memory usage to avoid costly failures from under-provisioning, which results in significant memory wastage.We propose KS+, a method that predicts a task’s memory consumption over time depending on its inputs. For this, KS+ dynamically segments the task execution and predicts the memory required for each segment. Our experimental evaluation shows an average reduction in memory wastage of 38% compared to the best-performing state-of-the-art baseline for two real-world workflows from the popular nf-core repository.

Correlating detailed local resource utilization data with the high-level concepts of distributed scientific workflow systems eventually causing it is challenging. When running a large-scale scientific data analysis workflow across a distributed execution environment, we want to analyze its I/O behaviour to identify potential bottlenecks. Since tasks are assigned to any available nodes, local resource usage on a node does not directly show which tasks are causing it. We acquire resource usage profiles of the involved nodes to link them to the individual workflow tasks. This is done by properly associating low-level trace metadata with high-level task information from log files and job management systems like Kubernetes. This information helps identifying areas of the workflow on a logical task level where improvements can make the biggest impact.

The pitfalls of centralized social networks, such as Facebook and Twitter/X, have led to concerns about control, transparency, and accountability. Decentralized social networks have emerged as a result with the goal of empowering users. These decentralized approaches come with their own trade-offs, and therefore multiple architectures exist. In this paper, we conduct the first large-scale analysis of Bluesky, a prominent decentralized microblogging platform. In contrast to alternative approaches (e.g. Mastodon), Bluesky decomposes and opens the key functions of the platform into subcomponents that can be provided by third party stakeholders. We collect a comprehensive dataset covering all the key elements of Bluesky, study user activity and assess the diversity of providers for each sub-components.

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.

Small satellites enable many important applications for both economic and scientific purposes. Many of these applications are inherently data-centric and rely on large amounts of high-resolution satellite imagery to be delivered in a timely manner. However, communicating this data to Earth is challenging due to intermittent connectivity, high packet losses, low data rates, and similar issues. Therefore, efficient onboard prioritization and data processing are essential for future satellite missions. Machine learning methods, such as deep neural networks, are very suitable for such prioritization, as they are already used extensively for satellite imagery processing and they can be deployed onboard of satellites. However, updating them to support new classification requirements when the satellite is already in orbit is difficult, as often multiple passes are required to complete model transmission due to the communication challenges. To cope with this issue, we propose a progressive transmission mechanism for model updates, which leverages vector quantization and arithmetic coding. Our mechanism allows to achieve high accuracies even with partially updated models. Evaluation results show that our mechanism significantly outperforms other less optimized transmission schemes.

In the realm of High-Performance Computing (HPC), the disparity between I/O and processing capabilities can hinder application performance. One source of this I/O gap is the latency of the devices, especially in case of distributed filesystems (DFS). Prior solutions to minimize such negative effects (e.g. using asynchronous operations or separate I/O-Threads), have to be considered when designing an application. This work aims to execute these operations asynchronously for existing applications without necessitating modifications or even recompilation. To attain this objective, we introduce the Lazy Read approach, which alters how the operating system handles read requests. This modification enables the application to start multiple reads or do CPU work in parallel instead of waiting for completion of the read operation. For evaluation, we implemented the Lazy Read approach in the Linux kernel. We then showed that the approach adds overhead, however, multiple read operations can be executed in parallel. In our specific evaluation the maximum possible speedup was achieved, reducing the execution time by up to 50%.

Scientific workflows have received increasing interest and are used in many scientific fields to gather, analyze, and process significant amounts of data. However, their tasks are usually treated as black boxes, and their behavior remains unconsidered for resource allocations, which can lead to subpar resource allocations with typical scheduling. Although not done yet, it should be possible to observe such tasks, learn their behavior, and use this knowledge to improve future executions. As workflows and their tasks are often executed multiple times on a massive scale, even a slight improvement per execution may save hours of execution time and significant amounts of energy.To achieve this goal, we develop an innovative approach to model task executions and predict resource usage. The prediction is embedded in a feedback loop to repeatedly improve the models and to closely track workflow executions to make predictions and resource allocations accurate.

Interplanetary Filesystem (IPFS) is one of the largest peer-to-peer filesystems in operation. The network is the default storage layer for Web3 and is being presented as a solution to the centralization of the web. In this paper, we present a large-scale, multi-modal measurement study of the IPFS network. We analyze the topology, the traffic, the content providers and the entry points from the classical Internet. Our measurements show significant centralization in the IPFS network and a high share of nodes hosted in the cloud. We also shed light on the main stakeholders in the ecosystem. We discuss key challenges that might disrupt continuing efforts to decentralize the Web and highlight multiple properties that are creating pressures toward centralization.

Small satellites are widely used today as cost effective means to perform Earth observation and other tasks that generate large amounts of high-dimensional data, such as multispectral imagery. These satellites typically operate in low earth orbit, which poses significant challenges for data transmission due to short contact times with ground stations, low bandwidth, and high packet loss probabilities. In this paper, we introduce JSCC-SAT, which applies joint source-and-channel coding using neural networks to provide efficient and robust transmission of compressed image data for satellite applications. We evaluate our mechanism against traditional transmission schemes with separate source and channel coding and demonstrate that it outperforms the existing approaches when applied to Earth observation data of the Sentinel-2 mission.

The growing heterogenous ecosystem of networked consumer devices such as smart meters or IoT-connected appliances such as air conditioners is difficult to secure, unlike the utility side of the grid which can be defended effectively through rigorous IT security measures such as isolated control networks. In this paper, we consider a crisis scenario in which an attacker compromises a large number of consumer-side devices and modulates their electrical power to destabilize the grid and cause an electrical outage [9, 26, 27, 47, 50, 55].

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 de scribe the input da ta 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.

Decentralized data storage systems like the Interplanetary Filesystem (IPFS) are becoming increasingly popular, e. g., as a data layer in blockchain applications and for sharing content in a censorship-resistant manner. In IPFS, data is hosted by an open set of nodes and data requests are broadcast to connected peers in addition to being routed via a distributed hash table (DHT). In this paper, we present a passive monitoring methodology that exploits this design for obtaining data requests from a significant and upscalable portion of nodes. Using an implementation of our approach for the IPFS network and data collected over a period of fifteen months, we demonstrate how our methodology enables profound insights into, among other things: the size of the IPFS network, activity levels and structure, and content popularity distributions. We furthermore present that our methodology can be abused for attacks on users’ privacy. For example, we were able to identify and successfully surveil the IPFS nodes corresponding to public IPFS/HTTP gateways. We give a detailed analysis of the mechanics and reasons behind implied privacy threats and discuss possible countermeasures.

Small satellites networks are characterized by intermittent connectivity and rare communication opportunities between satellites and ground stations. To overcome this problem, inter-satellite links can be used and predictability of satellite movements can be exploited to plan the communication in advance. Mathematical optimization techniques are typically applied to create such communication plans. However, the resulting optimization problems are often large and therefore problematic to solve. To handle them more efficiently, we propose a novel decomposition approach based on Benders decomposition. To the best of our knowledge, we are the first to apply Benders decomposition in the context of satellite networks. Our evaluation results show that our approach considerably outperforms the state-of-the-art solution.

Abstract Federated Byzantine Agreement Systems (FBASs) are a fascinating new paradigm in the context of consensus protocols. Originally proposed for powering the Stellar payment network, FBASs can instantiate Byzantine quorum systems without requiring out-of-band agreement on a common set of validators; every node is free to decide for itself with whom it requires agreement. Sybil-resistant and yet energy-efficient consensus protocols can therefore be built upon FBASs, and the “decentrality” possible with the FBAS paradigm might be sufficient to reduce the use of environmentally unsustainable proof-of-work protocols. In this paper, we first demonstrate how the robustness of individual FBASs can be determined, by precisely determining their safety and liveness buffers and therefore enabling a comparison with threshold-based quorum systems. Using simulations and example node configuration strategies, we then empirically investigate the hypothesis that while FBASs can be bootstrapped in a bottom-up fashion from individual preferences, strategic considerations should additionally be applied by node operators in order to arrive at FBASs that are robust and amenable to monitoring. Finally, we investigate the reported “open-membership” property of FBASs. We observe that an often small group of nodes is exclusively relevant for determining liveness buffers and prove that membership in this top tier is conditional on the approval by current top tier nodes if maintaining safety is a core requirement.

Many solutions have been proposed to improve manual contact tracing for infectious diseases through automation. Privacy is crucial for the deployment of such a system as it greatly influences adoption. Approaches for digital contact tracing like Google Apple Exposure Notification (GAEN) protect the privacy of users by decentralizing risk scoring. But GAEN leaks information about diagnosed users as ephemeral pseudonyms are broadcast to everyone. To combat deanonymisation based on the time of encounter while providing extensive risk scoring functionality we propose to use a private set intersection (PSI) protocol based on garbled circuits. Using oblivious programmable pseudo random functions PSI (OPPRF-PSI) , we implement our solution CERTAIN which leaks no information to querying users other than one risk score for each of the last 14 days representing their risk of infection. We implement payload inclusion for OPPRF-PSI and evaluate the efficiency and performance of different risk scoring mechanisms on an Android device.

To combat the ongoing Covid-19 pandemic, many new ways have been proposed on how to automate the process of finding infected people, also called contact tracing . A special focus was put on preserving the privacy of users. Bluetooth Low Energy as base technology has the most promising properties, so this survey focuses on automated contact tracing techniques using Bluetooth Low Energy. We define multiple classes of methods and identify two major groups: systems that rely on a server for finding new infections and systems that distribute this process. Existing approaches are systematically classified regarding security and privacy criteria.

Contact tracing is a promising approach to combat the COVID-19 pandemic. Various systems have been proposed to automatise the process. Many designs rely heavily on a centralised server or reveal significant amounts of private data to health authorities. We propose CAUDHT, a decentralized peer-to-peer system for contact tracing. The central health authority can focus on providing and operating tests for the disease while contact tracing is done by the system's users themselves. We use a distributed hash table to build a decentral messaging system for infected patients and their contacts. With blind signatures, we ensure that messages about infections are authentic and unchanged. A strong privacy focus enables data integrity, confidentiality, and privacy.

In multi-satellite formations for Earth observation, downlink capacity is a key bottleneck in data acquisition. As measurement data from different satellites of the same mission is often correlated to some degree, modern distributed source coding techniques are a promising approach for increasing downlink coding efficiency, thereby improving the overall mission data acquisition capability. Distributed arithmetic coding (DAC) has been extensively studied but has been applied primarily either on synthetic multi-source pseudo-random data or in single-source scenarios with real measurement data. In order to apply DAC to independent sources' measurement data of unknown entropy and correlation, we propose an adaptive DAC decoder and different use-case specific fitness functions. Using real world data from the MagSat mission, we demonstrate that correlations in physical measurement data can be exploited to increase coding efficiency of downlinks in multi-satellite Earth observation missions.

It is nowadays common in all kinds of traffic-maritime, air, road-that vehicles use beaconing systems to broadcast their position and other navigational data to nearby vehicles to improve situational awareness. MAC protocols for the beaconing use case face special challenges: protocol overhead has a significant performance impact due to small packet size, fair and frequent access to the channel is required, and the broadcast nature limits the applicability of hand shakes. We propose LAMA, a MAC protocol for position awareness beaconing that is based on the locally shared position information. Our contention-free approach uses neither handshakes nor forwarding of state, thereby requiring only small constant protocol overhead per beacon and scaling well with large neighbor counts. Yet, it successfully suppresses interference, including hidden-terminal-type, while maximizing channel utilization through coordinated spatial reuse. In a quantitative evaluation using the ns-3 discrete event simulator we compare LAMA against SO-TDMA, the MAC protocol used in the maritime automatic identification system. We find that LAMA outperforms SO-TDMA with respect to several metrics in synthetic random topologies as well as in scenarios based on real vessel traffic traces.

The protocol evaluation is an integral part of network protocol design. From the perspective of experimental design, discrete event simulations constitute a middle ground between analytical protocol evaluation and testbeds. They allow precise control of otherwise external influences while supporting more detailed protocol models than analytical evaluations. Compared to testbeds, a major restriction is that existing protocols require a separate implementation in the discrete event model. Creating this implementation model may cause differences between the protocol's simulator-based model and the native implementation, invalidating simulation results. We propose a novel architecture to evaluate unmodified, binary protocol implementations in the state-of-the-art discrete event simulators by utilizing the operating system's system call barrier. Notably, our approach does not affect discrete simulation properties, such as repeatability, and it does not require the native protocol implementation's source code. The evaluation results using existing network protocols show the feasibility of our approach in combination with the ns-3 simulator core. We show that our approach more closely resembles realistic protocol performance when compared to simulator-based protocol models. Moreover, our approach performs better than existing solutions for more realistic protocol simulations.

Many privacy-enhancing architectures require an anonymous communication channel as a component to acquire tokens or other material from a centralized authority without revealing the requester's identity. While anonymous routing overlays like Tor can provide the required anonymity for TCPlike data streams, it increases dependence on third party infrastructure, and widespread use may overload the Tor network's current capacity. Therefore, we propose to use requests in the Domain Name System as an anonymization layer by forwarding packets over public DNS resolvers operating in recursive mode. In contrast to existing approaches, we focus on UDP datagram transport. As an example use case, we implemented a system for anonymous issuance of blind signatures. We evaluate our domaintailored solution against existing approaches like Tor and analyze its performance and privacy guarantees. Evaluation results show that the achievable data rate using the DNS-based anonymous channel is sufficient for many practical use cases, and that the provided anonymity is comparable to that provided by the Tor network.

Today, the most widely distributed type of firewalls are software firewalls, running as applications on standard systems. Dedicated networking hardware like SDN switches has also been used to implement firewall functionality, although their native classification capabilities are restricted to relatively simple checks, such as subnet tests. However, native SDN hardware can satisfy high performance requirements that would be challenging for standard software firewalls. This motivates to build a hybrid combination of fast SDN hardware with a standard software firewall. Our approach directly offloads simple rule policies instead of flows to the SDN switch, in order to exploit the limited storage capacity more efficiently. An effective packet diversion algorithm based on header space analysis avoids expensive communication with the back-end. This way, it can achieve the throughput of a native SDN switch while still being able to resort decisions to the full extent of a software firewall. Our evaluation demonstrates a 23-fold classification performance increase over a standard software firewall.

Recent works have shown that the wireless link between two nodes is neither a shared secret nor unforgeable by an adversary-questioning the effectiveness of many physical layer security mechanisms. In this work, we argue that physical layer security is still valuable when multiple nodes combine their measurements of overheard adversarial transmissions. This assumption is applicable to, e.g., industrial wireless networks, where nodes are spatially close and wireless links are unique. First, we design a novel physical layer adversary detection algorithm particularly suited for the scenario of multiple sensing nodes. Second, we evaluate our algorithm with real-world measurements, which show that already four detecting nodes are sufficient to significantly increase the detection performance.

Packet classification is a core feature needed in firewalls, SDN switches, and QoS routers. Current research to accelerate the classification with hardware employing Field-programmable Gate Arrays (FPGAs) created a variety of approaches, with significant differences in terms of hardware resource requirements, memory usage, configuration update time, and power dissipation. However, there is no optimal, universal method for classification at link rate, due to inherent conflicts between large generic circuits with high resource consumption, and optimized circuits with limited versatility. Thus, current implementations have different trade-offs in terms of memory usage, resource requirements, power consumption, and flexibility. As a new approach to tackle this challenge, we present a hybrid concept that combines an highly optimized configuration-specialized and thus energy- and resource-efficient classification circuit with a generic matching circuit whose configuration can be updated quickly. The combined circuit can thus support reasonably fast configuration updates, has a low power dissipation, and can process network packets at link rate.

The trend towards smart factories necessitates protocols for wireless transmission of production process information. Protocols must transfer process information timely even with temporary wireless interference on the factory floor and regardless of machines' location in the network topology. TANDEM is a topology-independent wireless multi-hop network protocol that implements in-network prioritization to provide a useful, early approximation of process information. Tailored to often required time series data transmission, TANDEM applies a time-frequency transformation to prioritize more valuable information parts, decouples multi-hop communication steps to improve robustness, and uses a periodic acknowledgment overhearing mechanism to save transmissions. We analytically discuss reliability and fairness aspects and show using simulations that TANDEM's prioritization is beneficial to obtain an early preview of sensor information even in large factories. In addition, we implement TANDEM on industrial grade hardware and provide a performance evaluation.

Bonus point systems are widely used for rewarding customer loyalty in both traditional and electronic commerce scenarios. Merchants aim to additionally increase revenue by targeted advertising using customer data. At the same time, customers wish to protect their privacy of what they purchase. Common systems neither provide sufficient protection of privacy, nor protect the customers' identities. Anonymity makes it significantly harder to properly resolve claims of fraudulent transactions, because identities are not recorded for any purpose after a transaction has been committed. We propose an anonymous bonus system based on digital payment systems using blind cryptographic signatures. We implement a protocol extension that allows proof of bonus point possession in case of reported misbehavior, and we identify the implications on security, privacy, and performance of our proposals. Our results show that we can resolve these cases of fraud within the system without losing functionality in the bonus point protocol.

In recent years, many visions for hitherto considered-futuristic computing applications gained momentum. The vision of smart factories is one example for these trends, which all share a common requirement: the prolific dissemination of sensor information. As wireless communication in smart factories needs to cope with harsh environments, the amount of sensor information produced by sources will likely surpass the communication channel's available capacity. This discrepancy calls for efficient communication and filtering protocols, as well as compression mechanisms, as a foundation for dependable applications. We propose such a compression algorithm that is lossless and tailored towards the requirements of the manufacturing industry. Our algorithm employs a two-step stochastic model that uses lossy compression to extract an approximation from the signal and a separate noise model to accommodate the remaining error. Evaluation results validate that our algorithm achieves better compression rates than existing approaches for several types of real world sensor data from the industry.

Firewalls, key components for secured network infrastructures, are faced with two different kinds of challenges: first, they must be fast enough to classify network packets at line speed, and second, their packet processing capabilities should be versatile in order to support complex filtering policies. Unfortunately, most existing classification systems do not qualify equally well for both requirements: systems built on special-purpose hardware are fast, but limited in their filtering functionality. In contrast, software filters provide powerful matching semantics, but struggle to meet line speed. This motivates the combination of parallel, yet complexity-limited specialized circuitry with a slower, but versatile software firewall. The key challenge in such a design arises from the dependencies between classification rules due to their relative priorities within the rule set: complex rules requiring software-based processing may be interleaved at arbitrary positions between those where hardware processing is feasible. Therefore, we discuss approaches for partitioning and transforming rule sets for hybrid packet processing. As a result, we propose HyPaFilter+, a hybrid classification system consisting of an FPGA-based hardware matcher and a Linux netfilter firewall, which provides a simple, yet effective hardware/software packet shunting algorithm. Our evaluation shows up to 30-fold throughput gains over software packet processing.

With network traffic rates continuously growing, security systems like firewalls are facing increasing challenges to process incoming packets at line speed without sacrificing protection. Accordingly, specialized hardware firewalls are increasingly used in high-speed environments. Hardware solutions, though, are inherently limited in terms of the complexity of the policies they can implement, often forcing users to choose between throughput and comprehensive analysis. On the contrary, complex rules typically constitute only a small fraction of the rule set. This motivates the combination of massively parallel, yet complexity-limited specialized circuitry with a slower, but semantically powerful software firewall. The key challenge in such a design arises from the dependencies between classification rules due to their relative priorities within the rule set: complex rules requiring software-based processing may be interleaved at arbitrary positions between those where hardware processing is feasible. We therefore discuss approaches for partitioning and transforming rule sets for hybrid packet processing, and propose HyPaFilter, a hybrid classification system based on tailored circuitry on an FPGA as an accelerator for a Linux netfilter firewall. Our evaluation demonstrates 30-fold performance gains in comparison to software-only processing.

Network packet classification is a key functionality for packet filters and firewalls, and its performance is crucial for such systems to maintain a high packet throughput under heavy load situations. However, many existing packet filters employ slow classification algorithms which cannot provide the required lookup performance due to slow rule set traversal. In this work, we address this problem by providing a novel rule set transformation strategy called Minflate which combines the advantages of existing orthogonal transformation schemes by first minimizing a source rule set and then encoding decision trees into the minimized rule set. Our results show that the Minflate-generated rule sets are both small and can in many cases be traversed faster than rule sets transformed by existing techniques in isolation.

Today's science is more and more driven by collecting and evaluating increasing amounts of data. Utilizing Scientific Workflows is one suitable method how to organize processing pipelines for this purpose. In this work, we show that performance improvements on the execution of existing workflows can be achieved, if the conditions for starting selected tasks with certain data access characteristics are loosened. We provide a scheme how to identify eligible tasks in a given workflow and demonstrate a technique how an earlier start of tasks can be realized in Pegasus WMS by transforming the workflow DAG and by using a wrapper around the task executable during runtime. Our implemented wrapper handles the reading data accesses for task instances so that existing original workflows can be executed without the need to modify them. We evaluate our approach in simulations and experiments on real distributed computing resources, and are able to observe performance improvements for the Montage workflow by a significant reduction of total execution time.

While location management is a key component of cellular networks, it is also a major privacy issue: location management empowers the network operator to track users. In today's public and scientific discussion, the centralized storage of location data is mostly taken as a fact, and users are expected to trust the network operator. With ANOTEL we present a novel, clean-slate approach of location management in cellular networks that challenges this assumption. The design is able to route calls to users who move through cellular networks, without violating their location privacy.

Packet classification-the matching of packet headers against a predefined rule set-is a crucial functionality of firewalls, intrusion detection systems, and SDN switches. Most existing classification algorithms trade setup time for classification speed-that is, the packet classification is fast, but the transformation of rules set into the corresponding search data structure takes a considerable amount of time. This preprocessing time, however, poses a significant challenge for systems where rule sets can often change. Hence, these systems often use slow classification algorithms that support frequent rule set updates, which drastically limits their achievable throughput. In this work, we present a novel algorithmic technique which is able to "upgrade" an arbitrary existing classification algorithm to support fast updates, while still providing high lookup performance. Our evaluation demonstrates that our proposed technique exceeds the matching performance of existing dynamically updatable algorithms by an order of magnitude while providing the same level of update responsiveness.

Random linear network coding simplifies routing decisions, improves throughput, and increases tolerance against packet loss. A substantial limitation, however, is delay: decoding requires as many independent linear combinations as data blocks. Hierarchical network coding purportedly solves this delay problem. It introduces layers to decode prioritized data blocks early, which may benefit video streaming applications or applications for sensor information collection. While hierarchical network coding reduces decoding delays, it introduces significant space complexity and additional decoding time. We propose a decoding algorithm that manages all prioritization layers in a joint decoder matrix. Analytical evaluation and performance measurements show that we maintain prioritization benefits without increased space complexity and improve decoding performance. With memory requirements independent of the number of layers, our algorithm facilitates more fine-grained prioritization layers to further the benefits of hierarchical network coding.

Network packet classification plays a pivotal role in packet-switched networks; it is at the heart of many functions including firewalling, QoS routing, and OpenFlow-based switching. However, packet classification is a hard problem, as packets must be classified within a short time frame. Existing classification techniques use sophisticated data structures which are traversed by generic search algorithms - that is, the algorithm is static, while the data structure reflects the configuration of the classifier. In this paper, we propose to break up the strict separation between data structure and algorithm by specializing the algorithm's implementation on the specific classification rules. We demonstrate the feasibility of our approach by introducing JitVector, which builds upon the well-known bit vector classification algorithm, but generates instance-specific machine code at runtime. In our evaluation, which also includes an integration into the OpenFlow Reference Switch, we show that JitVector achieves significant performance gains over an equivalent generic search scheme.

Network packet classification is performed by a wide variety of network devices, like routers or firewalls. Accordingly, researchers have put great efforts in the development of fast packet classification algorithms. However, despite the fact that such approaches have been around for over a decade, most classification systems used in practice still rely on the slow linear search approach. In this work, we propose a methodology that enables linear search-based systems with jump semantics to take advantage of the superior matching performance of decision tree algorithms, without the need to touch the underlying system implementation. By performing source-to-source transformations on packet classification rule sets, we encode decision trees inside of the modified rule sets in order to guide and tweak the originally linear matching process. We implement this in a proof-of-concept tool which transforms Linux iptables firewall rule sets. Our evaluation demonstrates that throughput performance boosts of one order of magnitude and more are possible - without changing the semantics of the rule set, and without any modifications to the matching engine.

Many companies and institutions employ packet filter firewalls in order to effectively regulate network traffic. Unfortunately, the constant growth of network bandwidth makes the task of matching packet headers against potentially large rulesets more difficult, and prohibits the sole use of entirely software-based firewalls which cannot cope with such huge amounts of traffic. Instead, high-speed firewalls are often implemented in ASICs which offer a high degree of parallelism, many opportunities for operation pipelining, and low-latency access to network data. However, due to their static nature, ASICs must provide generic filtering circuitry that is hardly able to take full advantage of firewall ruleset properties, thus leading to a waste of hardware resources.

The process of matching the header fields of network packets against a set of rules is a performance critical task of firewalls. Software-based solutions have no chance to keep pace with the ever-growing data rates in high-speed networks. However, specialized filtering hardware is costly because complex logic is required in order to be able to apply arbitrary rulesets to a packet stream. By adapting the implemented logic to the specific firewall ruleset, FPGAs allow for much more specifically tailored and thus more efficient processing than ruleset-independent circuits in an ASIC. We present MPFC, a method to generate customized firewall circuits in the form of synthesizable VHDL code for FPGA configuration. The highly parallel MPFC circuits achieve a deterministic throughput of one packet per clock cycle, can be operated at high clock frequencies, and provide orders of magnitudes shorter processing latencies than previous work in this direction.

Peer reviewing is the key ingredient of evaluating the quality of scientific work. Based on the review scores assigned by the individual reviewers to the submissions, program committees of conferences and journal editors decide which papers to accept for publication and which to reject. However, some reviewers may be more rigorous than others, they may be biased one way or the other, and they often have highly subjective preferences over the papers they review. Moreover, each reviewer usually has only a very local view, as he or she evaluates only a small fraction of the submissions. Despite all these shortcomings, the review scores obtained need to be aggregrated in order to globally rank all submissions and to make the acceptance/rejection decision. A common method is to simply take the average of each submission's review scores, possibly weighted by the reviewers' confidence levels. Unfortunately, the global ranking thus produced often suffers a certain unfairness, as the reviewers' biases and limitations are not taken into account. We propose a method for calibrating the scores of reviewers that are potentially biased and blindfolded by having only partial information. Our method uses a maximum likelihood estimator, which estimates both the bias of each individual reviewer and the unknown "ideal" score of each submission. This yields a quadratic program whose solution transforms the individual review scores into calibrated, globally comparable scores. We argue why our method results in a fairer and more reasonable global ranking than simply taking the average of scores. To show its usefulness, we test our method empirically using real-world data.

Tor is one of the most popular network anonymization services. With increasing popularity, however, Tor is also faced with increasing load. Mechanisms for handling congestion and fairness in anonymization networks, where user privacy is of greatest significance, are not yet well understood. Thus current designs leave a lot to be desired: gross unfairness and largely suboptimal performance can be observed. In this paper, we focus on fairness aspects in the Tor network. We first show that interactions of multiple scheduling mechanisms in the current Tor design cause heavily unfair resource allocations to users. Subsequently, we develop a fairness model based on max-min fairness that takes the specifics of Tor into account. This leads us to a re-design of Tor's scheduling. We implement the new design in conjunction with a congestion feedback mechanism named N23, which has recently been proposed to be used in Tor. Our scheduling approach overcomes the unfairness problems which are exhibited by today's Tor implementation, and by Tor with N23 as well. It achieves global max-min fairness and thus a fair resource allocation despite selfish end-users.

On today's high-speed backbone network links, measuring per-flow traffic information has become very challenging. Maintaining exact per-flow packet counters on OC-192 or OC-768 links is not practically feasible due to computational and cost constrains. Packet sampling as implemented in today's routers results in large approximation errors. Here, we present Probabilistic Multiplicity Counting (PMC), a novel data structure that is capable of accounting traffic per flow probabilistically. The PMC algorithm is very simple and highly parallelizable, and therefore allows for efficient implementations in software and hardware. At the same time, it provides very accurate traffic statistics. We evaluate PMC with both artificial and real-world traffic data, demonstrating that it outperforms other approaches.

In TCP-based overlay applications, the TCP connections of one peer typically share one physical Internet link. Using real-world experiments, we demonstrate that this can lead to undesirable interactions, causing significant throughput loss. We argue that such effects should be taken into account in the design of overlay networks, and identify readily deployable countermeasures. In a first step, we show that with existing operating system QoS functionality some relief is possible. Yet, this alone is not fully effective if peers communicate bidirectionally, due to piggybacked ACKs. We propose to separate bidirectionally used overlay links into two independent TCP connections, and demonstrate the effectiveness of this strategy.

WiFi sharing communities offer an attractive alternative to commercial hot-spots and cellular networks for users who seek Internet access while being away from home. The growing number of mobile devices with Internet usage capabilities has formed a basis for the establishment and growth of several such communities. However, critical security problems exist in all current WiFi sharing approaches because they employ a central operator. We show how to overcome these issues by decentralizing the community's organization structure and present radioActive WiFi, a system that implements this approach.

During the analysis of packet log files from network experiments, the question arises which received packet belongs to which of the potentially many binary identical send events. We discuss this send-receive correlation problem for networks with local broadcast media. We can prove that assigning send and receive events is an NP-complete problem. However, there is a solution algorithm that is exponential only in the number of nodes; if the number of network nodes is fixed, its complexity is polynomial.

Access point sharing communities are an attractive solution for low-cost global mobile Internet access. However, existing communities all require a central authority in form of a certificate authority or an authentication server. In this paper, we show how such a community network can be created without a central entity. We introduce a mechanism called remote station approval; a host can use it to offer Internet access without requiring a central instance and without the risk of legal responsibility for the guests' traffic. We discuss how our system preserves the anonymity of its users and present a prototype implementation and some first performance figures.

Searching for witnesses in case of road accidents is a challenging task for the police and involved persons. In this paper, we propose a mechanism that helps to find witnesses. Our solution preserves the potential witnesses' anonymity and gives them a free hand to decide whether to step forward as a witness or not. We analyze the performance of our application using a realistic model of a German city.

Search indices for distributed information are an important building block for many mobile, decentralized applications. However, the network load caused by nodes registering their information in the index can become quite significant especially in mobile networks like vehicular ad-hoc networks (VANETs), where nodes need to register periodically due to their mobility. Our goal in this paper is to encode this registration information as efficiently as possible. For this purpose, we first analytically study hash keys and Bloom filters as alternative encoding types of nodes' registration data. It turns out that in many situations, Bloom filters are more bandwidth efficient than hash keys, and that they are even more efficient when nodes cooperate. Many such forms of cooperation are conceivable; we define one specific scheme targeting metropolitan VANETs. Our accompanying simulation study shows that this first algorithm already allows for large bandwidth savings in content registration, compared to the alternative of registering hash keys.

The distribution of dynamic information from many sources to many destinations is a key challenge for VANET applications such as cooperative traffic information management or decentralized parking guidance systems. In order for these systems to remain scalable it has been proposed to aggregate the information within the network as it travels from the sources to the destinations. However, so far it has remained unclear by what amount the aggregation scheme needs to reduce the original data in order to be considered scalable. In this paper we prove formally that any suitable aggregation scheme must reduce the bandwidth at which information about an area at distance d is provided to the cars asymptotically faster than 1/d2. Furthermore, we constructively show that this bound is tight: for any arbitrary ε>0, there exists a scalable aggregation scheme that reduces information asymptotically like 1/d(2+ε).

The recent work on COPE by Katti et al. demonstrates a practical application of network coding to wireless multihop networks. We note, however, that the opportunistic nature of COPE leaves it at the mercy of higher and lower layer protocols to create coding opportunities spontaneously. In this paper, we go one step beyond COPE's opportunism and study how to create coding opportunities in a more deterministic, yet still practical way. We start from the insight that in two-way traffic the existence of coding opportunities can be guaranteed through carefully co-ordinated packet scheduling, and establish general properties of protocols that are able to achieve this. We then propose Near-Optimal Coordinated Coding (noCoCo), a cross-layer scheme that integrates per-hop packet scheduling, network coding, and congestion control in a novel way. Extensive simulations show that noCoCo significantly outperforms standard non-coding approaches as well as COPE in terms of network throughput, delay and transmission overhead.

In mobile ad-hoc networks, the multicast paradigm is of central importance. It can help to save scarce medium bandwidth if packets are to be delivered to multiple destinations. We consider the problem of congestion control for multicast traffic in wireless multihop networks. We propose to apply a congestion control concept which is tailored to the very special properties of the wireless multihop medium: implicit hop-by-hop congestion control. The idea, so far only having been considered for unicast traffic, is here generalized to multicast. We implement it in the Backpressure Multicast Congestion Control (BMCC) protocol, with a focus on how to realize it in combination with geographic multicast routing in the Scalable Position-Based Multicast (SPBM) protocol. Our evaluation points out a number of highly desirable properties of the proposed scheme. In particular, it achieves and maintains high throughput and high packet delivery ratios at low packet latencies, even in the presence of significant network load.

A central problem in today's Internet is unsolicited bulk email: spam. The SMTP protocol lacks a mechanism for verifying the source of a message, and respective protocol extensions are still far from becoming standard. Content-based automatic spam filters are thus often used, and simple origin-based filtering techniques like black- and whitelists are also very common. In this paper we first analyze the retry behavior of Internet mail sources upon delivery errors. We then propose a novel spam filtering approach, founded on the results of this analysis. It is based on the observed reaction of a mail source host to temporary errors. Furthermore, evaluation results from the production use of our filter on real mail servers are given, underlining its very good performance in practice.

In the context of vehicular ad-hoc networks (VANETs), a number of highly promising convenience applications have been proposed. These include collecting and distributing information on the traffic situation, distributed monitoring of road and weather conditions, and finding available parking places in a distributed, cooperative manner. Unfortunately, all of these applications face major problems when a VANET is used as a means to distribute the required information. In particular a large number of vehicles needs to be equipped with dedicated VANET technology before these applications can provide a useful service. Even if customers were willing to purchase a system which is not immediately useful, it would still take quite some time until the required density of equipped cars is reached. In contrast, affordable always-on mobile Internet access is already mainstream. Such Internet connectivity could be used to build the proposed applications in a different fashion: by using peer-to-peer communication, essentially creating a peer-to-peer network of cars sharing traffic information. This allows to overcome the limitations of VANETs, while it preserves their key benefits of decentralization and robustness. In this paper, we describe the technical challenges that arise from such an approach, point out relevant research directions, and outline possible starting points for solutions.

The search for free parking places is a promising application for vehicular ad hoc networks (VANETs). In order to guide drivers to a free parking place at their destination, it is necessary to estimate the occupancy state of the parking lots within the destination area at time of arrival. In this paper, we present a model to predict parking lot occupancy based on information exchanged among vehicles. In particular, our model takes the age of received parking lot information and the time needed to arrive at a certain parking lot into account and estimates the future parking situation at time of arrival. It is based on queueing theory and uses a continuous-time homogeneous Markov model. We have evaluated the model in a simulation study based on a detailed model of the city of Brunswick, Germany.

Abstract Congestion control is a key problem in mobile ad hoc networks. The standard congestion control mechanism of the Transmission Control Protocol (TCP) is not able to handle the special properties of a shared wireless multi‐hop channel well. In particular, the frequent changes of the network topology and the shared nature of the wireless channel pose significant challenges. Many approaches have been proposed to overcome these difficulties. In this paper, we give an overview over existing proposals, explain their key ideas and show their interrelations. Copyright © 2007 John Wiley & Sons, Ltd.

Discrete-event network simulation is a major tool for the research and development of mobile ad-hoc networks (MANETs). These simulations are used for debugging, teaching, understanding, and performance-evaluating MANET protocols. For the first three tasks, visualization of the processes occurring in the simulated network is crucial for verification and credibility of the generated results. Working with the popular network simulator ns-2, we have not yet found a visualization toolkit capable of reading native ns-2 trace files and providing means to change the evaluated parameters without changing the visualization software. Thus, we developed Huginn, a software providing an intuitive way to visualize simulation properties and to determine how they should be displayed without the need of programming. In addition, Huginn has a 3D interface allowing an improved exploitation of the (human) user's perceptual system. It helps to handle the significant cognitive load associated with the mental reconstruction of simulated network processes. Besides presenting the software interface and architecture, we describe algorithmic solutions that might be of a more general interest for similar problems.