arXiv subjects

Volker Markl

Publications and source records attributed to Volker Markl.

31 records · Page 2Linked to original sources

arXiv2021

BigEarthNet-MM: A Large Scale Multi-Modal Multi-Label Benchmark Archive for Remote Sensing Image Classification and Retrieval

This paper presents the multi-modal BigEarthNet (BigEarthNet-MM) benchmark archive made up of 590,326 pairs of Sentinel-1 and Sentinel-2 image patches to support the deep learning (DL) studies in multi-modal multi-label remote sensing (RS) image retrieval and classification. Each pair of patches in BigEarthNet-MM is annotated with multi-labels provided by the CORINE Land Cover (CLC) map of 2018 based on its thematically most detailed Level-3 class nomenclature. Our initial research demonstrates that some CLC classes are challenging to be accurately described by only considering (single-date) BigEarthNet-MM images. In this paper, we also introduce an alternative class-nomenclature as an evolution of the original CLC labels to address this problem. This is achieved by interpreting and arranging the CLC Level-3 nomenclature based on the properties of BigEarthNet-MM images in a new nomenclature of 19 classes. In our experiments, we show the potential of BigEarthNet-MM for multi-modal multi-label image retrieval and classification problems by considering several state-of-the-art DL models. We also demonstrate that the DL models trained from scratch on BigEarthNet-MM outperform those pre-trained on ImageNet, especially in relation to some complex classes, including agriculture and other vegetated and natural environments. We make all the data and the DL models publicly available at https://bigearth.net, offering an important resource to support studies on multi-modal image scene classification and retrieval problems in RS.

cs.CV↗

arXiv2021

The Case for Distance-Bounded Spatial Approximations

Spatial approximations have been traditionally used in spatial databases to accelerate the processing of complex geometric operations. However, approximations are typically only used in a first filtering step to determine a set of candidate spatial objects that may fulfill the query condition. To provide accurate results, the exact geometries of the candidate objects are tested against the query condition, which is typically an expensive operation. Nevertheless, many emerging applications (e.g., visualization tools) require interactive responses, while only needing approximate results. Besides, real-world geospatial data is inherently imprecise, which makes exact data processing unnecessary. Given the uncertainty associated with spatial data and the relaxed precision requirements of many applications, this vision paper advocates for approximate spatial data processing techniques that omit exact geometric tests and provide final answers solely on the basis of (fine-grained) approximations. Thanks to recent hardware advances, this vision can be realized today. Furthermore, our approximate techniques employ a distance-based error bound, i.e., a bound on the maximum spatial distance between false (or missing) and exact results which is crucial for meaningful analyses. This bound allows to control the precision of the approximation and trade accuracy for performance.

cs.DB↗

arXiv2020

Agora: A Unified Asset Ecosystem Going Beyond Marketplaces and Cloud Services

Data, algorithms, and compute/storage infrastructure are key assets that drive data science and artificial intelligence applications. As providing all these assets requires a huge investment, data science and artificial intelligence technologies are currently dominated by a small number of providers who can afford these investments. This leads to lock-in effects and hinders features that require a flexible exchange of assets among users. In this vision paper, we present Agora, a unified asset ecosystem. The Agora system provides the technical infrastructure that allows for offering and using data and algorithms, as well as physical infrastructure components. Agora is designed as an open ecosystem of asset marketplaces and provides to a broad audience not only data but the entire data value chain (including computational resources and human expertise). Agora (i) leverages a fine-grained exchange of assets, (ii) allows for combining assets to novel applications, and (iii) flexibly executes such applications on available resources. As a result, Agora overcomes lock-in effects and removes entry barriers for new asset providers. In contrast to existing data management systems, Agora operates in a heavily decentralized and dynamic environment: Data, algorithms, and even compute resources are dynamically created, modified, and removed by different stakeholders. Agora presents novel research directions for the data management community as a whole: It requires to combine our traditional expertise in scalable data processing and management with infrastructure provisioning as well as economic and application aspects of data, algorithms, and infrastructure.

cs.DB↗

arXiv2020

Dynamic Parameter Allocation in Parameter Servers

To keep up with increasing dataset sizes and model complexity, distributed training has become a necessity for large machine learning tasks. Parameter servers ease the implementation of distributed parameter management---a key concern in distributed training---, but can induce severe communication overhead. To reduce communication overhead, distributed machine learning algorithms use techniques to increase parameter access locality (PAL), achieving up to linear speed-ups. We found that existing parameter servers provide only limited support for PAL techniques, however, and therefore prevent efficient training. In this paper, we explore whether and to what extent PAL techniques can be supported, and whether such support is beneficial. We propose to integrate dynamic parameter allocation into parameter servers, describe an efficient implementation of such a parameter server called Lapse, and experimentally compare its performance to existing parameter servers across a number of machine learning tasks. We found that Lapse provides near-linear scaling and can be orders of magnitude faster than existing parameter servers.

cs.LG↗

arXiv2020

The NebulaStream Platform: Data and Application Management for the Internet of Things

The Internet of Things (IoT) presents a novel computing architecture for data management: a distributed, highly dynamic, and heterogeneous environment of massive scale. Applications for the IoT introduce new challenges for integrating the concepts of fog and cloud computing as well as sensor networks in one unified environment. In this paper, we highlight these major challenges and outline how existing systems handle them. To address these challenges, we introduce the NebulaStream platform, a general purpose, endto-end data management system for the IoT. NebulaStream addresses the heterogeneity and distribution of compute and data, supports diverse data and programming models going beyond relational algebra, deals with potentially unreliable communication, and enables constant evolution under continuous operation. In our evaluation, we demonstrate the effectiveness of our approach by providing early results on partial aspects.

cs.DC↗

arXiv2019

SENSE: Scalable Data Acquisition from Distributed Sensors with Guaranteed Time Coherence

Data analysis in the Internet of Things (IoT) requires us to combine event streams from a huge amount of sensors. This combination (join) of events is usually based on the time stamps associated with the events. We address two challenges in environments which acquire and join events in the IoT: First, due to the growing number of sensors, we are facing the performance limits of central joins with respect to throughput, latency, and network utilization. Second, in the IoT, diverse sensor nodes are operated by different organizations and use different time synchronization techniques. Thus, events with the same timestamps are not necessarily recorded at the exact same time and joined data tuples have an unknown time incoherence. This can cause undetected failures, such as false correlations and wrong predictions. We present SENSE, a system for scalable data acquisition from distributed sensors. SENSE introduces time coherence measures as a fundamental data characteristic in addition to common time synchronization techniques. The time coherence of a data tuple is the time span in which all values contained in the tuple have been read from sensors. We explore concepts and algorithms to quantify and optimize time coherence and show that SENSE scales to thousands of sensors, operates efficiently under latency and coherence constraints, and adapts to changing network conditions.

cs.DB↗

arXiv2019

Benchmarking Distributed Stream Data Processing Systems

The need for scalable and efficient stream analysis has led to the development of many open-source streaming data processing systems (SDPSs) with highly diverging capabilities and performance characteristics. While first initiatives try to compare the systems for simple workloads, there is a clear gap of detailed analyses of the systems' performance characteristics. In this paper, we propose a framework for benchmarking distributed stream processing engines. We use our suite to evaluate the performance of three widely used SDPSs in detail, namely Apache Storm, Apache Spark, and Apache Flink. Our evaluation focuses in particular on measuring the throughput and latency of windowed operations, which are the basic type of operations in stream analytics. For this benchmark, we design workloads based on real-life, industrial use-cases inspired by the online gaming industry. The contribution of our work is threefold. First, we give a definition of latency and throughput for stateful operators. Second, we carefully separate the system under test and driver, in order to correctly represent the open world model of typical stream processing deployments and can, therefore, measure system performance under realistic conditions. Third, we build the first benchmarking framework to define and test the sustainable performance of streaming systems. Our detailed evaluation highlights the individual characteristics and use-cases of each system.

cs.DB↗

arXiv2019

BigEarthNet: A Large-Scale Benchmark Archive For Remote Sensing Image Understanding

This paper presents the BigEarthNet that is a new large-scale multi-label Sentinel-2 benchmark archive. The BigEarthNet consists of 590,326 Sentinel-2 image patches, each of which is a section of i) 120x120 pixels for 10m bands; ii) 60x60 pixels for 20m bands; and iii) 20x20 pixels for 60m bands. Unlike most of the existing archives, each image patch is annotated by multiple land-cover classes (i.e., multi-labels) that are provided from the CORINE Land Cover database of the year 2018 (CLC 2018). The BigEarthNet is significantly larger than the existing archives in remote sensing (RS) and thus is much more convenient to be used as a training source in the context of deep learning. This paper first addresses the limitations of the existing archives and then describes the properties of the BigEarthNet. Experimental results obtained in the framework of RS image scene classification problems show that a shallow Convolutional Neural Network (CNN) architecture trained on the BigEarthNet provides much higher accuracy compared to a state-of-the-art CNN model pre-trained on the ImageNet (which is a very popular large-scale benchmark archive in computer vision). The BigEarthNet opens up promising directions to advance operational RS applications and research in massive Sentinel-2 image archives.

cs.CV↗

arXiv2018

Labyrinth: Compiling Imperative Control Flow to Parallel Dataflows

Parallel dataflow systems have become a standard technology for large-scale data analytics. Complex data analysis programs in areas such as machine learning and graph analytics often involve control flow, i.e., iterations and branching. Therefore, systems for advanced analytics should include control flow constructs that are efficient and easy to use. A natural approach is to provide imperative control flow constructs similar to those of mainstream programming languages: while-loops, if-statements, and mutable variables, whose values can change between iteration steps. However, current parallel dataflow systems execute programs written using imperative control flow constructs by launching a separate dataflow job after every control flow decision (e.g., for every step of a loop). The performance of this approach is suboptimal, because (a) launching a dataflow job incurs scheduling overhead; and (b) it prevents certain optimizations across iteration steps. In this paper, we introduce Labyrinth, a method to compile programs written using imperative control flow constructs to a single dataflow job, which executes the whole program, including all iteration steps. This way, we achieve both efficiency and ease of use. We also conduct an experimental evaluation, which shows that Labyrinth has orders of magnitude smaller per-iteration-step overhead than launching new dataflow jobs, and also allows for significant optimizations across iteration steps.

cs.DC↗

arXiv2018

A Survey of State Management in Big Data Processing Systems

State management and its use in diverse applications varies widely across big data processing systems. This is evident in both the research literature and existing systems, such as Apache Flink, Apache Samza, Apache Spark, and Apache Storm. Given the pivotal role that state management plays in various use cases, in this survey, we present some of the most important uses of state as an enabler, discuss the alternative approaches used to handle and implement state, propose a taxonomy to capture the many facets of state management, and highlight new research directions. Our aim is to provide insight into disparate state management techniques, motivate others to pursue research in this area, and draw attention to some open problems.

cs.DB↗

arXiv2017

Generating Custom Code for Efficient Query Execution on Heterogeneous Processors

Processor manufacturers build increasingly specialized processors to mitigate the effects of the power wall to deliver improved performance. Currently, database engines are manually optimized for each processor: A costly and error prone process. In this paper, we propose concepts to enable the database engine to perform per-processor optimization automatically. Our core idea is to create variants of generated code and to learn a fast variant for each processor. We create variants by modifying parallelization strategies, specializing data structures, and applying different code transformations. Our experimental results show that the performance of variants may diverge up to two orders of magnitude. Therefore, we need to generate custom code for each processor to achieve peak performance. We show that our approach finds a fast custom variant for multi-core CPUs, GPUs, and MICs.

cs.DB↗

arXiv2017

Sharing Hash Codes for Multiple Purposes

Locality sensitive hashing (LSH) is a powerful tool for sublinear-time approximate nearest neighbor search, and a variety of hashing schemes have been proposed for different dissimilarity measures. However, hash codes significantly depend on the dissimilarity, which prohibits users from adjusting the dissimilarity at query time. In this paper, we propose {multiple purpose LSH (mp-LSH) which shares the hash codes for different dissimilarities. mp-LSH supports L2, cosine, and inner product dissimilarities, and their corresponding weighted sums, where the weights can be adjusted at query time. It also allows us to modify the importance of pre-defined groups of features. Thus, mp-LSH enables us, for example, to retrieve similar items to a query with the user preference taken into account, to find a similar material to a query with some properties (stability, utility, etc.) optimized, and to turn on or off a part of multi-modal information (brightness, color, audio, text, etc.) in image/video retrieval. We theoretically and empirically analyze the performance of three variants of mp-LSH, and demonstrate their usefulness on real-world data sets.

stat.ML↗

arXiv2012

Spinning Fast Iterative Data Flows

Parallel dataflow systems are a central part of most analytic pipelines for big data. The iterative nature of many analysis and machine learning algorithms, however, is still a challenge for current systems. While certain types of bulk iterative algorithms are supported by novel dataflow frameworks, these systems cannot exploit computational dependencies present in many algorithms, such as graph algorithms. As a result, these algorithms are inefficiently executed and have led to specialized systems based on other paradigms, such as message passing or shared memory. We propose a method to integrate incremental iterations, a form of workset iterations, with parallel dataflows. After showing how to integrate bulk iterations into a dataflow system and its optimizer, we present an extension to the programming model for incremental iterations. The extension alleviates for the lack of mutable state in dataflows and allows for exploiting the sparse computational dependencies inherent in many iterative algorithms. The evaluation of a prototypical implementation shows that those aspects lead to up to two orders of magnitude speedup in algorithm runtime, when exploited. In our experiments, the improved dataflow system is highly competitive with specialized systems while maintaining a transparent and unified dataflow abstraction.

cs.DB↗