CoChat
Sign Up
Cited by 17Advanced Database Systems and QueriesCloud Computing and Resource ManagementParallel Computing and Optimization Techniques

The DOI matches its Crossref record.

Quantitative Impact Evaluation of an Abstraction Layer for Data Stream Processing Systems

Hesse, Guenter, Matthies, Christoph, Glass, Kelvin, Huegle, Johannes, Uflacker, Matthias·2019·2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS)
DOI 10.1109/icdcs.2019.00138
Open this paper in CoChat — free account

Abstract

With the demand to process ever-growing data volumes, a variety of new data stream processing frameworks have been developed. Moving an implementation from one such system to another, e.g., for performance reasons, requires adapting existing applications to new interfaces. Apache Beam addresses these high substitution costs by providing an abstraction layer that enables executing programs on any of the supported streaming frameworks. In this paper, we present a novel benchmark architecture for comparing the performance impact of using Apache Beam on three streaming frameworks: Apache Spark Streaming, Apache Flink, and Apache Apex. We find significant performance penalties when using Apache Beam for application development in the surveyed systems. Overall, usage of Apache Beam for the examined streaming applications caused a high variance of query execution times with a slowdown of up to a factor of 58 compared to queries developed without the abstraction layer. All developed benchmark artifacts are publicly available to ensure reproducible results.

Overview

Integrity signals

Registry

Crossref · 10.1109/icdcs.2019.00138

Cites

Statistics

Recomputing p-values needs the paper’s full text.

Verification

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Using Apache Beam as an abstraction layer comes at the cost of significantly higher resource requirements regardless of the use case.

    “Using Apache Beam as an abstraction layer still comes at the cost of significantly higher resource requirements regardless of the use case.”

    From Benchmarking scalability of stream processing frameworks deployed as microservices in the cloud (2024) · cites this paper as Hesse et al., 2019

  • The performance overhead of using Apache Beam as an abstraction layer has also been observed in related research.

    “The performance overhead of using Apache Beam as an abstraction layer has also been observed in related research (Hesse et al., 2019).”

    From Benchmarking scalability of stream processing frameworks deployed as microservices in the cloud (2024) · cites this paper as Hesse et al., 2019

  • Spæren (2021) discovered unnecessary serialization and deserialization between operators and introduced the FasterCopy option, which disables these copy operations.

    “They discovered unnecessary serialization and deserialization between operators and introduced the FasterCopy option, which disables these copy operations.”

    From Benchmarking scalability of stream processing frameworks deployed as microservices in the cloud (2024) · cites this paper as Hesse et al. (2019)

  • Previous research found that using Apache Beam as an abstraction layer comes with a significant negative impact on performance.

    “Previous research found that using Apache Beam as an abstraction layer comes with a significant negative impact on performance (Hesse et al., 2019).”

    From Benchmarking scalability of stream processing frameworks deployed as microservices in the cloud (2024) · cites this paper as Hesse et al., 2019

Citations

Cites

No references could be resolved yet.

Research works printed but not linked: 68

  1. 0, Apache Flink
  2. 0, Apache Apex Documentation - Application Developer Guide
  3. kreps, 2011, Kafka: a Distributed Messaging System for Log Processing, Proc International Workshop on Networking Meets Databases NetDB, 1
  4. dunning, 2016, Streaming Architecture New Designs Using Apache Kafka and Mapr Streams
  5. 10.1007/978-3-319-54334-5_8 · DOI
  6. 0, HDFS Architecture
  7. 0, Apache Hadoop Yarn
  8. 0, Running Spark on YARN
  9. 0, Flink - YARN Setup
  10. 10.1145/2723372.2742790 · DOI
  11. 10.14778/3137765.3137770 · DOI · Paper page
  12. 10.1007/s00778-003-0095-z · DOI
  13. arasu, 0, Linear Road: A Stream Data Management Benchmark, Proceedings of International Conference on Very Large Data Bases, 480
  14. 0, Apache Storm
  15. zaharia, 0, Discrctized Streams: Fault-Tolerant Streaming Computation at Scale, ACM SIGOPS Symposium on Operating Systems Principles SOSP, 423
  16. 0, NEXMark Benchmark
  17. zaharia, 0, Resilient Distributed Datasets: A Fault-Tolerant Abstraction for In-Memory Cluster Computing, Proc of USENIX Symposium on Networked Systems Design & Implementation, 15
  18. tucker, 0, NEXMark - A Benchmark for Queries over Data Streams DRAFT
  19. 0, Nexmark benchmark suite
  20. 0, Apache Hadoop
  21. marcu, 0, Spark versus Flink: Understanding Performance in Big Data Analytics Frameworks, Proceedings IEEE International Conference on Cluster Computing CLUSTR, 433
  22. 10.1109/GLOCOM.2016.7841533 · DOI
  23. 0, Apache Beam Overview
  24. 10.1109/ICDE.2005.1 · DOI
  25. 0, Flink - Distributed Runtime Environment
  26. 0, Apache Spark - Cluster Mode Overview
  27. 0, Spark Streaming Programming Guide
  28. 10.1145/2523616.2523633 · DOI · Paper page
  29. hindman, 0, Mesos: A Platform for Fine-Grained Resource Sharing in the Data Center, Proc of USENIX Symposium on Networked Systems Design & Implementation
  30. 10.1109/HOTI.2014.15 · DOI
  31. 10.1145/2806777.2809955 · DOI
  32. 10.1007/s00778-004-0147-z · DOI
  33. arasu, 2003, STREAM: The Stanford Stream Data Manager, IEEE Data Eng Bull, 26, 19
  34. 10.1109/IPDPSW.2016.141 · DOI
  35. 0, SAP HANA Smart Data Streaming Developer Guide
  36. 0, KSQL and Kafka Streams
  37. 0, Overview of STREAMS
  38. 0, Oracle® CEP CQL Language Reference 11g Release 1 (11 1 1)
  39. 10.14778/1454159.1454179 · DOI
  40. 0, StreamingMedia
  41. 10.1145/3183713.3190662 · DOI
  42. 10.14778/2824032.2824076 · DOI · Paper page
  43. 0, Beam Capability Matrix
  44. 0, Mirror of Apache Apex core
  45. 0, Apache Gearpump
  46. 0, MapReduce Tutorial
  47. 0, What is Samza?
  48. 0, Alibaba jstorm
  49. 0, IBM Streams
  50. 0, CLOUD DATAFLOW - Simplified stream and batch data processing with equal reliability and expressiveness
  51. 0, Cloud Dataflow Apache Beam and you
  52. 10.1109/ICPADS.2015.106 · DOI
  53. carbone, 2015, Apache Flink™: Stream and Batch Processing in a Single Engine, IEEE Data Eng Bull, 38, 28
  54. 10.24251/HICSS.2017.592 · DOI
  55. 0, Apache Apex
  56. 10.1145/2934664 · DOI · Paper page
  57. 0, Apache Beam Programming Guide
  58. 0, Apache Beam
  59. 10.14778/3229863.3229864 · DOI
  60. 0, Runner Authoring Guide
  61. 0, Spark configuration
  62. 0, Flink - Command-Line Interface
  63. 10.1109/ICDE.2018.00169 · DOI
  64. 0, Interface Context OperatorContext
  65. 0, AOL Search Query Logs
  66. 0, Mirror of Apache Spark
  67. 0, Documentation - Kafka 0 10 2 Documentation
  68. lu, 0, StreamBench: Towards Benchmarking Modern Distributed Stream Computing Frameworks, Proc IEEE/ACM International Conference on Utility and Cloud Computing UCC, 69

Cited by

Sorted by citations

Metadata

Authors
Hesse, G., Matthies, C., Glass, K., Huegle, J., & Uflacker, M.
Affiliations
  • Hasso Plattner Institute (DE)
  • Freie Universität Berlin (DE)
ORCID
Venue
2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS)
Publisher
IEEE
Published
2019-07
Pages
1381-1392
DOI
10.1109/icdcs.2019.00138
OpenAlex
W2964151964
Cited by
17
References
0
Citations by year
2026: 1 · 2025: 2 · 2024: 3 · 2023: 4 · 2022: 2 · 2021: 4 · 2020: 1
FWCI
1.18
Type
conference paper
Language
en
Open access
closed
Licence
Locations
IEEE International Conference on Distributed Computing Systems (ICDCS) (publishedVersion)
Topics
  • Physical Sciences › Computer Science › Information Systems › Advanced Database Systems and Queries
  • Physical Sciences › Computer Science › Information Systems › Cloud Computing and Resource Management
  • Physical Sciences › Computer Science › Hardware and Architecture › Parallel Computing and Optimization Techniques
Keywords
stream processing, abstraction layer, Apache Beam, performance impact, Spark Streaming, Apache Flink, query execution time

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review