Lauritz Thamsen

Kathleen West, Vasilis Bountris, Philipp Thamm et al.

Scientific workflows are widely used to process large quantities of data, leading to significant energy consumption and carbon emissions. To reduce this environmental impact, energy and carbon-aware scheduling approaches could be employed. However, such methods require runtime and energy predictions, which are typically only available for workflows that have been executed previously. Meanwhile, scientists may execute new or modified workflows, use workflows with different input data, or run them on alternative infrastructure. To address this critical gap, we propose Augur, a novel method to predict the energy consumption of scientific workflow tasks prior to execution. By efficiently profiling both the available cluster infrastructure and the workflow at hand, Augur is capable of predicting the overall energy consumption of the workflow with a median prediction error of $16.3\pm15.3\%$ compared to Ichnos, an energy estimation method that uses fitted power models, and $18.2\pm14.7\%$ compared to Intel RAPL, as observed in our experimental evaluation on public and private cloud infrastructure. Relying on only minimal historical execution data, Augur outperforms two state-of-the-art methods in predicting both task runtime and total workflow energy, providing a robust foundation for energy-efficient and carbon-aware scientific data analysis.

James Nurdin, Wei Liu, Richard Mccreadie et al.

Data analytics increasingly runs on distributed lakehouse systems, where platform operators must optimise monetary, resource, and environmental costs. Query Performance Prediction (QPP) helps to balance these costs and supports workload management techniques, such as adaptive resource scaling and low-carbon scheduling. However, runtimes in lakehouses can vary substantially, and the impact of runtime variance on QPP accuracy and workload orchestration has not previously been systematically studied for lakehouse systems. This paper addresses this gap by investigating the runtime variance observed for distributed lakehouse analytical queries and its impact on QPP. First, we quantify the run-to-run variance using Kubernetes deployments across three public clouds and one private cloud, spanning multiple database scales and three analytical benchmarks. Our results demonstrate that repeated executions of the same query can vary in runtime by nearly twofold. Second, we conduct a factor analysis study assessing key sources of this runtime variance such as data locality, co-tenant load, and caching effects. Third, we examine how variance influences state-of-the-art QPP models, revealing that addressing key sources of variance can reduce prediction error up to 80%. Finally, we demonstrate the downstream implications for low-carbon scheduling as an example of a workload management technique that relies on performance prediction, showing that accounting for runtime variance can lead to a significant reduction in carbon costs.

Houkun Zhu, Dominik Scheinert, Lauritz Thamsen et al.

Distributed file systems are widely used nowadays, yet using their default configurations is often not optimal. At the same time, tuning configuration parameters is typically challenging and time-consuming. It demands expertise and tuning operations can also be expensive. This is especially the case for static parameters, where changes take effect only after a restart of the system or workloads. We propose a novel approach, Magpie, which utilizes deep reinforcement learning to tune static parameters by strategically exploring and exploiting configuration parameter spaces. To boost the tuning of the static parameters, our method employs both server and client metrics of distributed file systems to understand the relationship between static parameters and performance. Our empirical evaluation results show that Magpie can noticeably improve the performance of the distributed file system Lustre, where our approach on average achieves 91.8% throughput gains against default configuration after tuning towards single performance indicator optimization, while it reaches 39.7% more throughput gains against the baseline.

Dominik Scheinert, Soeren Becker, Jonathan Bader et al.

Choosing a good resource configuration for big data analytics applications can be challenging, especially in cloud environments. Automated approaches are desirable as poor decisions can reduce performance and raise costs. The majority of existing automated approaches either build performance models from previous workload executions or conduct iterative resource configuration profiling until a near-optimal solution has been found. In doing so, they only obtain an implicit understanding of the underlying infrastructure, which is difficult to transfer to alternative infrastructures and, thus, profiling and modeling insights are not sustained beyond very specific situations. We present Perona, a novel approach to robust infrastructure fingerprinting for usage in the context of big data analytics. Perona employs common sets and configurations of benchmarking tools for target resources, so that resulting benchmark metrics are directly comparable and ranking is enabled. Insignificant benchmark metrics are discarded by learning a low-dimensional representation of the input metric vector, and previous benchmark executions are taken into consideration for context-awareness as well, allowing to detect resource degradation. We evaluate our approach both on data gathered from our own experiments as well as within related works for resource configuration optimization, demonstrating that Perona captures the characteristics from benchmark runs in a compact manner and produces representations that can be used directly.

Dominik Scheinert, Babak Sistani Zadeh Aghdam, Soeren Becker et al.

With increasingly more computation being shifted to the edge of the network, monitoring of critical infrastructures, such as intermediate processing nodes in autonomous driving, is further complicated due to the typically resource-constrained environments. In order to reduce the resource overhead on the network link imposed by monitoring, various methods have been discussed that either follow a filtering approach for data-emitting devices or conduct dynamic sampling based on employed prediction models. Still, existing methods are mainly requiring adaptive monitoring on edge devices, which demands device reconfigurations, utilizes additional resources, and limits the sophistication of employed models. In this paper, we propose a sampling-based and cloud-located approach that internally utilizes probabilistic forecasts and hence provides means of quantifying model uncertainties, which can be used for contextualized adaptations of sampling frequencies and consequently relieves constrained network resources. We evaluate our prototype implementation for the monitoring pipeline on a publicly available streaming dataset and demonstrate its positive impact on resource efficiency in a method comparison.

Dominik Scheinert, Philipp Wiesner, Thorsten Wittkopp et al.

Selecting the right resources for big data analytics jobs is hard because of the wide variety of configuration options like machine type and cluster size. As poor choices can have a significant impact on resource efficiency, cost, and energy usage, automated approaches are gaining popularity. Most existing methods rely on profiling recurring workloads to find near-optimal solutions over time. Due to the cold-start problem, this often leads to lengthy and costly profiling phases. However, big data analytics jobs across users can share many common properties: they often operate on similar infrastructure, using similar algorithms implemented in similar frameworks. The potential in sharing aggregated profiling runs to collaboratively address the cold start problem is largely unexplored. We present Karasu, an approach to more efficient resource configuration profiling that promotes data sharing among users working with similar infrastructures, frameworks, algorithms, or datasets. Karasu trains lightweight performance models using aggregated runtime information of collaborators and combines them into an ensemble method to exploit inherent knowledge of the configuration search space. Moreover, Karasu allows the optimization of multiple objectives simultaneously. Our evaluation is based on performance data from diverse workload executions in a public cloud environment. We show that Karasu is able to significantly boost existing methods in terms of performance, search time, and cost, even when few comparable profiling runs are available that share only partial common characteristics with the target job.

Philipp Thamm, Somayeh Mohammadi, Kathleen West et al.

Scientific workflows are pipelines of interdependent tasks. They are increasingly executed on shared Kubernetes clusters via workflow engines such as Nextflow. Their energy consumption matters for both cost and sustainability. It is necessary to examine and optimize workflow tasks individually, because they can be very heterogeneous. However, estimating task-level energy on clusters is difficult: Intel RAPL counters report only node-level energy, access to counters and host process information is typically restricted, and concurrent workloads introduce resource contention and measurement noise. We present Nf-PEAK, a containerized method to attribute CPU-package and DRAM energy to individual processes and Nextflow tasks. Nf-PEAK (i) identifies workflow pods, (ii) maps pods to host processes via cgroup metadata, (iii) samples RAPL and per-process performance counters, and (iv) applies a non-linear energy-credit model before aggregating results at task level. On a Kubernetes cluster, we evaluate three nf-core workflows under controlled co-located CPU load. Nf-PEAK reaches an average Mean Absolute Percentage Error of 6.6% in isolated runs and 10.9% when an unrelated workload saturates 8 of 32 hardware threads per node, and remains stable across 2, 3, 4, and 8 nodes. Compared to the state-of-the-art Kubernetes tool Kepler, Nf-PEAK yields lower error on average, particularly under co-located load.

Theviyanthan Krishnamohan, Lauritz Thamsen, Paul Harvey

The reliance of organisations on computer networks is enabled by network programmability, which is typically achieved through Service Function Chaining. These chains virtualise network functions, link them, and programmatically embed them on networking infrastructure. Optimal embedding of Service Function Chains is an NP-hard problem, with three sub-problems, chain composition, virtual network function embedding, and link embedding, that have to be optimised simultaneously, rather than sequentially, for optimal results. Genetic Algorithms have been employed for this, but existing approaches either do not optimise all three sub-problems or do not optimise all three sub-problems simultaneously. We propose a Genetic Algorithm-based approach called GENESIS, which evolves three sine-function-activated Neural Networks, and funnels their output to a Gaussian distribution and an A* algorithm to optimise all three sub-problems simultaneously. We evaluate GENESIS on an emulator across 48 different data centre scenarios and compare its performance to two state-of-the-art Genetic Algorithms and one greedy algorithm. GENESIS produces an optimal solution for 100% of the scenarios, whereas the second-best method optimises only 71% of the scenarios. Moreover, GENESIS is the fastest among all Genetic Algorithms, averaging 15.84 minutes, compared to an average of 38.62 minutes for the second-best Genetic Algorithm.

Philipp Wiesner, Ramin Khalili, Dennis Grinwald et al.

Federated Learning (FL) is an emerging machine learning technique that enables distributed model training across data silos or edge devices without data sharing. Yet, FL inevitably introduces inefficiencies compared to centralized model training, which will further increase the already high energy usage and associated carbon emissions of machine learning in the future. One idea to reduce FL's carbon footprint is to schedule training jobs based on the availability of renewable excess energy that can occur at certain times and places in the grid. However, in the presence of such volatile and unreliable resources, existing FL schedulers cannot always ensure fast, efficient, and fair training. We propose FedZero, an FL system that operates exclusively on renewable excess energy and spare capacity of compute infrastructure to effectively reduce a training's operational carbon emissions to zero. Using energy and load forecasts, FedZero leverages the spatio-temporal availability of excess resources by selecting clients for fast convergence and fair participation. Our evaluation, based on real solar and load traces, shows that FedZero converges significantly faster than existing approaches under the mentioned constraints while consuming less energy. Furthermore, it is robust to forecasting errors and scalable to tens of thousands of clients.

Jossekin Beilharz, Philipp Wiesner, Arne Boockmeyer et al.

The continuous testing of small changes to systems has proven to be useful and is widely adopted in the development of software systems. For this, software is tested in environments that are as close as possible to the production environments. When testing IoT systems, this approach is met with unique challenges that stem from the typically large scale of the deployments, heterogeneity of nodes, challenging network characteristics, and tight integration with the environment among others. IoT test environments present a possible solution to these challenges by emulating the nodes, networks, and possibly domain environments in which IoT applications can be executed. This paper gives an overview of the state of the art in IoT testing. We derive desirable characteristics of IoT test environments, compare 18 tools that can be used in this respect, and give a research outlook of future trends in this area.

Dominik Scheinert, Alireza Alamgiralem, Jonathan Bader et al.

With the growing amount of data, data processing workloads and the management of their resource usage becomes increasingly important. Since managing a dedicated infrastructure is in many situations infeasible or uneconomical, users progressively execute their respective workloads in the cloud. As the configuration of workloads and resources is often challenging, various methods have been proposed that either quickly profile towards a good configuration or determine one based on data from previous runs. Still, performance data to train such methods is often lacking and must be costly collected. In this paper, we propose a collaborative approach for sharing anonymized workload execution traces among users, mining them for general patterns, and exploiting clusters of historical workloads for future optimizations. We evaluate our prototype implementation for mining workload execution graphs on a publicly available trace dataset and demonstrate the predictive value of workload clusters determined through traces only.

Dominik Scheinert, Houkun Zhu, Lauritz Thamsen et al.

Distributed dataflow systems like Spark and Flink enable the use of clusters for scalable data analytics. While runtime prediction models can be used to initially select appropriate cluster resources given target runtimes, the actual runtime performance of dataflow jobs depends on several factors and varies over time. Yet, in many situations, dynamic scaling can be used to meet formulated runtime targets despite significant performance variance. This paper presents Enel, a novel dynamic scaling approach that uses message propagation on an attributed graph to model dataflow jobs and, thus, allows for deriving effective rescaling decisions. For this, Enel incorporates descriptive properties that capture the respective execution context, considers statistics from individual dataflow tasks, and propagates predictions through the job graph to eventually find an optimized new scale-out. Our evaluation of Enel with four iterative Spark jobs shows that our approach is able to identify effective rescaling actions, reacting for instance to node failures, and can be reused across different execution contexts.

Kordian Gontarska, Morgan Geldenhuys, Dominik Scheinert et al.

Distributed Stream Processing (DSP) systems enable processing large streams of continuous data to produce results in near to real time. They are an essential part of many data-intensive applications and analytics platforms. The rate at which events arrive at DSP systems can vary considerably over time, which may be due to trends, cyclic, and seasonal patterns within the data streams. A priori knowledge of incoming workloads enables proactive approaches to resource management and optimization tasks such as dynamic scaling, live migration of resources, and the tuning of configuration parameters during run-times, thus leading to a potentially better Quality of Service. In this paper we conduct a comprehensive evaluation of different load prediction techniques for DSP jobs. We identify three use-cases and formulate requirements for making load predictions specific to DSP jobs. Automatically optimized classical and Deep Learning methods are being evaluated on nine different datasets from typical DSP domains, i.e. the IoT, Web 2.0, and cluster monitoring. We compare model performance with respect to overall accuracy and training duration. Our results show that the Deep Learning methods provide the most accurate load predictions for the majority of the evaluated datasets.

Dominik Scheinert, Lauritz Thamsen, Houkun Zhu et al.

Distributed dataflow systems enable the use of clusters for scalable data analytics. However, selecting appropriate cluster resources for a processing job is often not straightforward. Performance models trained on historical executions of a concrete job are helpful in such situations, yet they are usually bound to a specific job execution context (e.g. node type, software versions, job parameters) due to the few considered input parameters. Even in case of slight context changes, such supportive models need to be retrained and cannot benefit from historical execution data from related contexts. This paper presents Bellamy, a novel modeling approach that combines scale-outs, dataset sizes, and runtimes with additional descriptive properties of a dataflow job. It is thereby able to capture the context of a job execution. Moreover, Bellamy is realizing a two-step modeling approach. First, a general model is trained on all the available data for a specific scalable analytics algorithm, hereby incorporating data from different contexts. Subsequently, the general model is optimized for the specific situation at hand, based on the available data for the concrete context. We evaluate our approach on two publicly available datasets consisting of execution data from various dataflow jobs carried out in different environments, showing that Bellamy outperforms state-of-the-art methods.

Kordian Gontarska, Weronika Wrazen, Jossekin Beilharz et al.

Cardiovascular diseases and heart failures in particular are the main cause of non-communicable disease mortality in the world. Constant patient monitoring enables better medical treatment as it allows practitioners to react on time and provide the appropriate treatment. Telemedicine can provide constant remote monitoring so patients can stay in their homes, only requiring medical sensing equipment and network connections. A limiting factor for telemedical centers is the amount of patients that can be monitored simultaneously. We aim to increase this amount by implementing a decision support system. This paper investigates a machine learning model to estimate a risk score based on patient vital parameters that allows sorting all cases every day to help practitioners focus their limited capacities on the most severe cases. The model we propose reaches an AUCROC of 0.84, whereas the baseline rule-based model reaches an AUCROC of 0.73. Our results indicate that the usage of deep learning to improve the efficiency of telemedical centers is feasible. This way more patients could benefit from better health-care through remote monitoring.

Dominik Scheinert, Alexander Acker, Lauritz Thamsen et al.

Operation and maintenance of large distributed cloud applications can quickly become unmanageably complex, putting human operators under immense stress when problems occur. Utilizing machine learning for identification and localization of anomalies in such systems supports human experts and enables fast mitigation. However, due to the various inter-dependencies of system components, anomalies do not only affect their origin but propagate through the distributed system. Taking this into account, we present Arvalus and its variant D-Arvalus, a neural graph transformation method that models system components as nodes and their dependencies and placement as edges to improve the identification and localization of anomalies. Given a series of metric KPIs, our method predicts the most likely system state - either normal or an anomaly class - and performs localization when an anomaly is detected. During our experiments, we simulate a distributed cloud application deployment and synthetically inject anomalies. The evaluation shows the generally good prediction performance of Arvalus and reveals the advantage of D-Arvalus which incorporates information about system component dependencies.

Felix Lorenz, Lauritz Thamsen, Andreas Wilke et al.

Especially in context of critical urban infrastructures, trust in IoT data is of utmost importance. While most technology stacks provide means for authentication and encryption of device-to-cloud traffic, there are currently no mechanisms to rule out physical tampering with an IoT device's sensors. Addressing this gap, we introduce a new method for extracting a hardware fingerprint of an IoT sensor which can be used for secret-free authentication. By comparing the fingerprint against reference measurements recorded prior to deployment, we can tell whether the sensing hardware connected to the IoT device has been changed by environmental effects or with malicious intent. Our approach exploits the characteristic behavior of analog circuits, which is revealed by applying a fixed-frequency alternating current to the sensor, while recording its output voltage. To demonstrate the general feasibility of our method, we apply it to four commercially available temperature sensors using laboratory equipment and evaluate the accuracy. The results indicate that with a sensible configuration of the two hyperparameters we can identify individual sensors with high probability, using only a few recordings from the target device.