search

TU-Berlin-DIMA / scotty-window-processor

This repository provides Scotty, a framework for efficient window aggregations for out-of-order Stream Processing.
https://tu-berlin-dima.github.io/scotty-window-processor/
Apache License 2.0
75 stars 23 forks source link
flink flink-stream-processing stream stream-processing windowing

readme

Scotty: Efficient Window Aggregation for out-of-order Stream Processing Build Status

This repository provides Scotty, a framework for efficient window aggregations for out-of-order Stream Processing.

Features:

Website

Resources:

Flink Integration Example:

// Instantiate Scotty window operator
KeyedScottyWindowOperator<Tuple, Tuple2<Integer, Integer>, Tuple2<Integer, Integer>> windowOperator =
new KeyedScottyWindowOperator<>(new SumWindowFunction());

// Add multiple windows to the same operator
windowOperator.addWindow(new TumblingWindow(WindowMeasure.Time, 1000));
windowOperator.addWindow(new SlidingWindow(WindowMeasure.Time, 1000, 5000));
windowOperator.addWindow(new SessionWindow(WindowMeasure.Time, 1000));

// Add operator to Flink job
stream.keyBy(0)
      .process(windowOperator)

More examples for Flink as well as demos for the other systems are provided in the demo folder.

Benchmark:

Throughput in comparison to the Flink standard window operator (Window Buckets) for Sliding Event-Time Windows:
We fix the window size to 60 seconds and modify the slide size. If the slide size gets smaller, Flink has to maintain a higher number of overlapping (concurrent) windows.

Throughput in comparison to Flink for concurrent Tumbling Windows:

Out-of-order Processing in Scotty:

A watermark with timestamp t indicates that no tuple with a timestamp lower than t will arrive. When a watermark arrives, Scotty outputs all window results of windows that ended before t.

However, some out-of-order tuples may arrive with a timestamp t' lower than the watermark t (t' <= t). For that case, the maxLateness indicates how long Scotty stores slices and their corresponding window aggregates. Scotty processes an out-of-order tuple as long as its in the allowed lateness, i.e. it has an timestamp t' that is bigger than watermark minus the maxLateness (t' > t - maxLateness). Then, Scotty outputs updated aggregates for the windows. The maxLateness can be adjusted with the method setMaxLateness of the slicingWindowOperator.

If an out-of-order tuple arrives outside the allowed lateness (with a timestamp t' < t - maxLateness - maxFixedWindowSize), it is discarded. MaxFixedWindowSize is the maximum size of window types which have fixed sizes (e.g. tumbling window or sliding window). For context-aware window types, the maxFixedWindowSize is 0.

If no watermark has arrived yet, an out-of-order tuple is outside the allowed lateness, when its timestamp is lower than the start timestamp of the first slice (t' < tsOfFirstSlice), or in case of context aware windows, the timestamp of the first tuple minus the maxlateness (t' < tsOfFirstTuple - maxLateness).

Roadmap:

We plan to extend our framework with the following features:

Setup:

The maven package is currently not publically available. Therefore we have to build it from source:

git clone git@github.com:TU-Berlin-DIMA/scotty-window-processor.git

mvn clean install

Then you can use the library in your maven project.

<dependency> 
 <groupId>stream.scotty</groupId>
 <artifactId>flink-connector</artifactId>
 <version>0.4</version>
</dependency>

Efficient Window Aggregation with General Stream Slicing at EDBT 2019

General Stream Slicing received the Best Paper Award at the 22nd International Conference on Extending Database Technology in March 2019.

Abstract:
Window aggregation is a core operation in data stream processing. Existing aggregation techniques focus on reducing latency, eliminating redundant computations, and minimizing memory usage. However, each technique operates under different assumptions with respect to workload characteristics such as properties of aggregation functions (e.g., invertible, associative), window types (e.g., sliding, sessions), windowing measures (e.g., time- or countbased), and stream (dis)order. Violating the assumptions of a technique can deem it unusable or drastically reduce its performance. In this paper, we present the first general stream slicing technique for window aggregation. General stream slicing automatically adapts to workload characteristics to improve performance without sacrificing its general applicability. As a prerequisite, we identify workload characteristics which affect the performance and applicability of aggregation techniques. Our experiments show that general stream slicing outperforms alternative concepts by up to one order of magnitude.

Acknowledgements: This work was funded by the EU project E2Data (780245), Delft Data Science, and the German Ministry for Education and Research as BBDC I (01IS14013A) and BBDC II (01IS18025A)

Scotty: Efficient Window Aggregation for out-of-order Stream Processing at ICDE 2018

Scotty was first published at the 34th IEEE International Conference on Data Engineering in April 2018.

Abstract:
Computing aggregates over windows is at the core of virtually every stream processing job. Typical stream processing applications involve overlapping windows and, therefore, cause redundant computations. Several techniques prevent this redundancy by sharing partial aggregates among windows. However, these techniques do not support out-of-order processing and session windows. Out-of-order processing is a key requirement to deal with delayed tuples in case of source failures such as temporary sensor outages. Session windows are widely used to separate different periods of user activity from each other. In this paper, we present Scotty, a high throughput operator for window discretization and aggregation. Scotty splits streams into non-overlapping slices and computes partial aggregates per slice. These partial aggregates are shared among all concurrent queries with arbitrary combinations of tumbling, sliding, and session windows. Scotty introduces the first slicing technique which (1) enables stream slicing for session windows in addition to tumbling and sliding windows and (2) processes out-of-order tuples efficiently. Our technique is generally applicable to a broad group of dataflow systems which use a unified batch and stream processing model. Our experiments show that we achieve a throughput an order of magnitude higher than alternative stateof-the-art solutions.

Acknowledgements: This work was supported by the EU projects Proteus (687691) and Streamline (688191), DFG Stratosphere (606902), and the German Ministry for Education and Research as BBDC (01IS14013A) and Software Campus (01IS12056).

Scotty: General and Efficient Open-source Window Aggregation for Stream Processing Systems at ACM TODS 2021

Abstract:
Window aggregation is a core operation in data stream processing. Existing aggregation techniques focus on reducing latency, eliminating redundant computations, or minimizing memory usage. However, each technique operates under different assumptions with respect to workload characteristics, such as properties of aggregation functions (e.g., invertible, associative), window types (e.g., sliding, sessions), windowing measures (e.g., time- or count-based), and stream (dis)order. In this article, we present Scotty, an efficient and general open-source operator for sliding-window aggregation in stream processing systems, such as Apache Flink, Apache Beam, Apache Samza, Apache Kafka, Apache Spark, and Apache Storm. One can easily extend Scotty with user-defined aggregation functions and window types. Scotty implements the concept of general stream slicing and derives workload characteristics from aggregation queries to improve performance without sacrificing its general applicability. We provide an in-depth view on the algorithms of the general stream slicing approach. Our experiments show that Scotty outperforms alternative solutions

Acknowledgements: This work was supported by the German Ministry for Education and Research as BIFOLD (01IS18025A and 01IS18037A), SFB 1404 FONDA, and the EU Horizon 2020 Opertus Mundi project (870228).

Benson et al.: Disco: Efficient Distributed Window Aggregation

Abstract:
Many business applications benefit from fast analysis of online data streams. Modern stream processing engines (SPEs) provide complex window types and user-defined aggregation functions to analyze streams. While SPEs run in central data centers, wireless sensors networks (WSNs) perform distributed aggregations close to the data sources, which is beneficial especially in modern IoT setups. However, WSNs support only basic aggregations and windows. To bridge the gap between complex central aggregations and simple distributed analysis, we propose Disco, a distributed complex window aggregation approach. Disco processes complex window types on multiple independent nodes while efficiently aggregating incoming data streams. Our evaluation shows that Disco’s throughput scales linearly with the number of nodes and that Disco already outperforms a centralized solution in a two-node setup. Furthermore, Disco reduces the network cost significantly compared to the centralized approach. Disco’s treelike topology handles thousands of nodes per level and scales to support future data-intensive streaming applications.

Verwiebe et al.: Survey of window types for aggregation in stream processing systems

Abstract:
In this paper, we present the first comprehensive survey of window types for stream processing systems which have been presented in research and commercial systems. We cover publications from the most relevant conferences, journals, and system whitepapers on stream processing, windowing, and window aggregation which have been published over the last 20 years. For each window type, we provide detailed specifications, formal notations, synonyms, and use-case examples. We classify each window type according to categories that have been proposed in literature and describe the out-of-order processing. In addition, we examine academic, commercial, and open-source systems with respect to the window types that they support. Our survey offers a comprehensive overview that may serve as a guideline for the development of stream processing systems, window aggregation techniques, and frameworks that support a variety of window types.

Acknowledgements: This work was funded by German Research Foundation (MA4662-5 and 410830482), German Federal Ministry for Education and Research as BIFOLD - Berlin Institute for the Foundations of Learning and Data (ref. 01IS18025A and ref. 01IS18037A) and Software Campus (01IS17052).

Verwiebe et al.: Algorithms for Windowed Aggregations and Joins on Distributed Stream Processing Systems

Abstract:
Window aggregations and windowed joins are central operators of modern real-time analytic workloads and significantly impact the performance of stream processing systems. This paper gives an overview of state-of-the-art research in this area conducted by the Berlin Institute for the Foundations of Learning and Data (BIFOLD) and the Technische Universität Berlin. To this end, we present different algorithms for efficiently processing windowed operators and discuss techniques for distributed stream processing. Recently, several approaches have leveraged modern hardware for windowed stream processing, which we will also include in this overview. Additionally, we describe the integration of windowed operators into various stream processing systems and diverse applications that use specialized window operations