arXiv subjects

Lauritz Thamsen

Publications and source records attributed to Lauritz Thamsen.

At least 55 records · Page 3Linked to original sources

arXiv2023

Perona: Robust Infrastructure Fingerprinting for Resource-Efficient Big Data Analytics

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.

cs.DC↗

arXiv2023

Probabilistic Time Series Forecasting for Adaptive Monitoring in Edge Computing Environments

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.

cs.DC↗

arXiv2023

Khaos: Dynamically Optimizing Checkpointing for Dependable Distributed Stream Processing

Distributed Stream Processing systems are becoming an increasingly essential part of Big Data processing platforms as users grow ever more reliant on their ability to provide fast access to new results. As such, making timely decisions based on these results is dependent on a system's ability to tolerate failure. Typically, these systems achieve fault tolerance and the ability to recover automatically from partial failures by implementing checkpoint and rollback recovery. However, owing to the statistical probability of partial failures occurring in these distributed environments and the variability of workloads upon which jobs are expected to operate, static configurations will often not meet Quality of Service constraints with low overhead. In this paper we present Khaos, a new approach which utilizes the parallel processing capabilities of virtual cloud automation technologies for the automatic runtime optimization of fault tolerance configurations in Distributed Stream Processing jobs. Our approach employs three subsequent phases which borrows from the principles of Chaos Engineering: establish the steady-state processing conditions, conduct experiments to better understand how the system performs under failure, and use this knowledge to continuously minimize Quality of Service violations. We implemented Khaos prototypically together with Apache Flink and demonstrate its usefulness experimentally.

cs.DC↗

arXiv2023

Get Your Memory Right: The Crispy Resource Allocation Assistant for Large-Scale Data Processing

Distributed dataflow systems like Apache Spark and Apache Hadoop enable data-parallel processing of large datasets on clusters. Yet, selecting appropriate computational resources for dataflow jobs -- that neither lead to bottlenecks nor to low resource utilization -- is often challenging, even for expert users such as data engineers. Further, existing automated approaches to resource selection rely on the assumption that a job is recurring to learn from previous runs or to warrant the cost of full test runs to learn from. However, this assumption often does not hold since many jobs are too unique. Therefore, we present Crispy, a method for optimizing data processing cluster configurations based on job profiling runs with small samples of the dataset on just a single machine. Crispy attempts to extrapolate the memory usage for the full dataset to then choose a cluster configuration with enough total memory. In our evaluation on a dataset with 1031 Spark and Hadoop jobs, we see a reduction of job execution costs by 56% compared to the baseline, while on average spending less than ten minutes on profiling runs per job on a consumer-grade laptop.

cs.DC↗

arXiv2022

Reshi: Recommending Resources for Scientific Workflow Tasks on Heterogeneous Infrastructures

Scientific workflows typically comprise a multitude of different processing steps which often are executed in parallel on different partitions of the input data. These executions, in turn, must be scheduled on the compute nodes of the computational infrastructure at hand. This assignment is complicated by the facts that (a) tasks typically have highly heterogeneous resource requirements and (b) in many infrastructures, compute nodes offer highly heterogeneous resources. In consequence, predictions of the runtime of a given task on a given node, as required by many scheduling algorithms, are often rather imprecise, which can lead to sub-optimal scheduling decisions. We propose Reshi, a method for recommending task-node assignments during workflow execution that can cope with heterogeneous tasks and heterogeneous nodes. Reshi approaches the problem as a regression task, where task-node pairs are modeled as feature vectors over the results of dedicated micro benchmarks and past task executions. Based on these features, Reshi trains a regression tree model to rank and recommend nodes for each ready-to-run task, which can be used as input to a scheduler. For our evaluation, we benchmarked 27 AWS machine types using three representative workflows. We compare Reshi's recommendations with three state-of-the-art schedulers. Our evaluation shows that Reshi outperforms HEFT by a mean makespan reduction of 7.18% and 18.01% assuming a mean task runtime prediction error of 15%.

cs.DC↗

arXiv2022

Cucumber: Renewable-Aware Admission Control for Delay-Tolerant Cloud and Edge Workloads

The growing electricity demand of cloud and edge computing increases operational costs and will soon have a considerable impact on the environment. A possible countermeasure is equipping IT infrastructure directly with on-site renewable energy sources. Yet, particularly smaller data centers may not be able to use all generated power directly at all times, while feeding it into the public grid or energy storage is often not an option. To maximize the usage of renewable excess energy, we propose Cucumber, an admission control policy that accepts delay-tolerant workloads only if they can be computed within their deadlines without the use of grid energy. Using probabilistic forecasting of computational load, energy consumption, and energy production, Cucumber can be configured towards more optimistic or conservative admission. We evaluate our approach on two scenarios using real solar production forecasts for Berlin, Mexico City, and Cape Town in a simulation environment. For scenarios where excess energy was actually available, our results show that Cucumber's default configuration achieves acceptance rates close to the optimal case and causes 97.0% of accepted workloads to be powered using excess energy, while more conservative admission results in 18.5% reduced acceptance at almost zero grid power usage.

cs.DC↗

arXiv2022

Magpie: Automatically Tuning Static Parameters for Distributed File Systems using Deep Reinforcement Learning

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.

cs.DC↗

arXiv2022

Phoebe: QoS-Aware Distributed Stream Processing through Anticipating Dynamic Workloads

Distributed Stream Processing systems have become an essential part of big data processing platforms. They are characterized by the high-throughput processing of near to real-time event streams with the goal of delivering low-latency results and thus enabling time-sensitive decision making. At the same time, results are expected to be consistent even in the presence of partial failures where exactly-once processing guarantees are required for correctness. Stream processing workloads are oftentimes dynamic in nature which makes static configurations highly inefficient as time goes by. Static resource allocations will almost certainly either negatively impact upon the Quality of Service and/or result in higher operational costs. In this paper we present Phoebe, a proactive approach to system auto-tuning for Distributed Stream Processing jobs executing on dynamic workloads. Our approach makes use of parallel profiling runs, QoS modeling, and runtime optimization to provide a general solution whereby configuration parameters are automatically tuned to ensure a stable service as well as alignment with recovery time Quality of Service targets. Phoebe makes use of Time Series Forecasting to gain an insight into future workload requirements thereby delivering scaling decisions which are accurate, long-lived, and reliable. Our experiments demonstrate that Phoebe is able to deliver a stable service while at the same time reducing resource over-provisioning.

cs.DC↗

arXiv2022

Collaborative Cluster Configuration for Distributed Data-Parallel Processing: A Research Overview

Many organizations routinely analyze large datasets using systems for distributed data-parallel processing and clusters of commodity resources. Yet, users need to configure adequate resources for their data processing jobs. This requires significant insights into expected job runtimes and scaling behavior, resource characteristics, input data distributions, and other factors. Unable to estimate performance accurately, users frequently overprovision resources for their jobs, leading to low resource utilization and high costs. In this paper, we present major building blocks towards a collaborative approach for optimization of data processing cluster configurations based on runtime data and performance models. We believe that runtime data can be shared and used for performance models across different execution contexts, significantly reducing the reliance on the recurrence of individual processing jobs or, else, dedicated job profiling. For this, we describe how the similarity of processing jobs and cluster infrastructures can be employed to combine suitable data points from local and global job executions into accurate performance models. Furthermore, we outline approaches to performance prediction via more context-aware and reusable models. Finally, we lay out how metrics from previous executions can be combined with runtime monitoring to effectively re-configure models and clusters dynamically.

cs.DC↗

arXiv2022

Lotaru: Locally Estimating Runtimes of Scientific Workflow Tasks in Heterogeneous Clusters

Many scientific workflow scheduling algorithms need to be informed about task runtimes a-priori to conduct efficient scheduling. In heterogeneous cluster infrastructures, this problem becomes aggravated because these runtimes are required for each task-node pair. Using historical data is often not feasible as logs are typically not retained indefinitely and workloads as well as infrastructure changes. In contrast, online methods, which predict task runtimes on specific nodes while the workflow is running, have to cope with the lack of example runs, especially during the start-up. In this paper, we present Lotaru, a novel online method for locally estimating task runtimes in scientific workflows on heterogeneous clusters. Lotaru first profiles all nodes of a cluster with a set of short-running and uniform microbenchmarks. Next, it runs the workflow to be scheduled on the user's local machine with drastically reduced data to determine important task characteristics. Based on these measurements, Lotaru learns a Bayesian linear regression model to predict a task's runtime given the input size and finally adjusts the predicted runtime specifically for each task-node pair in the cluster based on the micro-benchmark results. Due to its Bayesian approach, Lotaru can also compute robust uncertainty estimates and provides them as an input for advanced scheduling methods. Our evaluation with five real-world scientific workflows and different datasets shows that Lotaru significantly outperforms the baselines in terms of prediction errors for homogeneous and heterogeneous clusters.

cs.DC↗

arXiv2022

Differentiating Network Flows for Priority-Aware Scheduling of Incoming Packets in Real-Time IoT Systems

When IP-packet processing is unconditionally carried out on behalf of an operating system kernel thread, processing systems can experience overload in high incoming traffic scenarios. This is especially worrying for embedded real-time devices controlling their physical environment in industrial IoT scenarios and automotive systems. We propose an embedded real-time aware IP stack adaption with an early demultiplexing scheme for incoming packets and subsequent per-flow aperiodic scheduling. By instrumenting existing embedded IP stacks, rigid prioritization with minimal latency is deployed without the need of further task resources. Simple mitigation techniques can be applied to individual flows, causing hardly measurable overhead while at the same time protecting the system from overload conditions. Our IP stack adaption is able to reduce the low-priority packet processing time by over 86% compared to an unmodified stack. The network subsystem can thereby remain active at a 7x higher general traffic load before disabling the receive IRQ as a last resort to assure deadlines.

cs.NI↗

arXiv2022

SyncMesh: Improving Data Locality for Function-as-a-Service in Meshed Edge Networks

The increasing use of Internet of Things devices coincides with more communication and data movement in networks, which can exceed existing network capabilities. These devices often process sensor or user information, where data privacy and latency are a major concern. Therefore, traditional approaches like cloud computing do not fit well, yet new architectures such as edge computing address this gap. In addition, the Function-as-a-Service (FaaS) paradigm gains in prevalence as a workload execution platform, however the decoupling of storage results in further challenges for highly distributed edge environments. To address this, we propose SyncMesh, a system to manage, query, and transform data in a scalable and stateless manner by leveraging the capabilities of Function-as-a-Service and at the same time enabling data locality. Furthermore, we provide a prototypical implementation and evaluate it against established centralized and decentralized systems in regard to traffic usage and request times. The preliminary results indicate that SyncMesh is able to exonerate the network layer and accelerate the transmission of data to clients, while simultaneously improving local data processing.

cs.DC↗

arXiv2022

Training Data Reduction for Performance Models of Data Analytics Jobs in the Cloud

Distributed dataflow systems like Apache Flink and Apache Spark simplify processing large amounts of data on clusters in a data-parallel manner. However, choosing suitable cluster resources for distributed dataflow jobs in both type and number is difficult, especially for users who do not have access to previous performance metrics. One approach to overcoming this issue is to have users share runtime metrics to train context-aware performance models that help find a suitable configuration for the job at hand. A problem when sharing runtime data instead of trained models or model parameters is that the data size can grow substantially over time. This paper examines several clustering techniques to minimize training data size while keeping the associated performance models accurate. Our results indicate that efficiency gains in data transfer, storage, and model training can be achieved through training data reduction. In the evaluation of our solution on a dataset of runtime data from 930 unique distributed dataflow jobs, we observed that, on average, a 75% data reduction only increases prediction errors by one percentage point.

cs.DC↗

arXiv2022

Enel: Context-Aware Dynamic Scaling of Distributed Dataflow Jobs using Graph Propagation

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.

cs.DC↗

arXiv2022

Tarema: Adaptive Resource Allocation for Scalable Scientific Workflows in Heterogeneous Clusters

Scientific workflow management systems like Nextflow support large-scale data analysis by abstracting away the details of scientific workflows. In these systems, workflows consist of several abstract tasks, of which instances are run in parallel and transform input partitions into output partitions. Resource managers like Kubernetes execute such workflow tasks on cluster infrastructures. However, these resource managers only consider the number of CPUs and the amount of available memory when assigning tasks to resources; they do not consider hardware differences beyond these numbers, while computational speed and memory access rates can differ significantly. We propose Tarema, a system for allocating task instances to heterogeneous cluster resources during the execution of scalable scientific workflows. First, Tarema profiles the available infrastructure with a set of benchmark programs and groups cluster nodes with similar performance. Second, Tarema uses online monitoring data of tasks, assigning labels to tasks depending on their resource usage. Third, Tarema uses the node groups and task labels to dynamically assign task instances evenly to resources based on resource demand. Our evaluation of a prototype implementation for Kubernetes, using five real-world Nextflow workflows from the popular nf-core framework and two 15-node clusters consisting of different virtual machines, shows a mean reduction of isolated job runtimes by 19.8% compared to popular schedulers in widely-used resource managers and 4.54% compared to the heuristic SJFN, while providing a better cluster usage. Moreover, executing two long-running workflows in parallel and on restricted resources shows that Tarema is able to reduce the runtimes even more while providing a fair cluster usage.

cs.DC↗

arXiv2022

On the Potential of Execution Traces for Batch Processing Workload Optimization in Public Clouds

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.

cs.DC↗

arXiv2022

A Priority-Aware Multiqueue NIC Design

Low-level embedded systems are used to control cyber-phyiscal systems in industrial and autonomous applications. They need to meet hard real-time requirements as unanticipated controller delays on moving machines can have devastating effects. Modern developments such as the industrial Internet of Things and autonomous machines require these devices to connect to large IP networks. Since Network Interface Controllers (NICs) trigger interrupts for incoming packets, real-time embedded systems are subject to unpredictable preemptions when connected to such networks. In this work, we propose a priority-aware NIC design to moderate network-generated interrupts by mapping IP flows to processes and based on that, consolidates their packets into different queues. These queues apply priority-dependent interrupt moderation. First experimental evaluations show that 93% of interrupts can be saved leading to an 80% decrease of processing delay of critical tasks in the configurations investigated.

cs.NI↗

arXiv2021

Continuously Testing Distributed IoT Systems: An Overview of the State of the Art

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.

cs.DC↗