Performance evaluation of real-time stream processing systems for Internet of Things applications

Vikash,; Mishra, Lalita; Varma, Shirshu

doi:10.1016/j.future.2020.07.012

Cited by 30 publications

(7 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Likewise, there are no further studies evaluating scalability with a systematic approach as we do in this study. Vikash et al [88], Nasiri et al [76], Karakaya et al [75], and van Dongen and van den Poel [79] explicitly evaluate scalability, however, without testing different load intensities against different resource amounts in isolated experiments. Nasiri et al [76] conduct independent evaluations of scaling load and computing resources and, thus, address another aspect than our study.…”

Section: Related Work On Stream Processing Benchmarkingmentioning

confidence: 99%

Benchmarking Scalability of Stream Processing Frameworks Deployed as Event-Driven Microservices in the Cloud

Henning,

Hasselbring

2023

Preprint

View full text Add to dashboard Cite

Section: Related Work On Stream Processing Benchmarkingmentioning

confidence: 99%

Benchmarking Scalability of Stream Processing Frameworks Deployed as Event-Driven Microservices in the Cloud

Henning,

Hasselbring

2023

Preprint

View full text Add to dashboard Cite

“…As in most studies [2,15,19], both our metrics describe scalability as a function instead of a scalar. Sanders et al [16] highlight that scalability is a function as usually capacity does not grow at a constant rate with additional resources [2,7,19]. However, they remark that scalability can be measured as a scalar within a range.…”

Section: Scalability As a Functionmentioning

confidence: 99%

“…We observe that in experimental studies, evaluating scalability as a function of resources is more frequent than evaluations as a function of load [2,10,15,19]. An issue with those metrics is that they contradict common scalability definitions, which concern the system's behavior if subject to increasing load (see Section 2).…”

Section: Function Of Load Vs Function Of Resourcesmentioning

confidence: 99%

How to Measure Scalability of Distributed Stream Processing Engines?

Henning

Hasselbring

2021

Companion of the ACM/SPEC International Conference on Performance Engineering

View full text Add to dashboard Cite

Scalability is promoted as a key quality feature of modern big data stream processing engines. However, even though research made huge efforts to provide precise definitions and corresponding metrics for the term scalability, experimental scalability evaluations or benchmarks of stream processing engines apply different and inconsistent metrics. With this paper, we aim to establish general metrics for scalability of stream processing engines. Derived from common definitions of scalability in cloud computing, we propose two metrics: a load capacity function and a resource demand function. Both metrics relate provisioned resources and load intensities, while requiring specific service level objectives to be fulfilled. We show how these metrics can be employed for scalability benchmarking and discuss their advantages in comparison to other metrics, used for stream processing engines and other software systems. CCS CONCEPTS• General and reference → Metrics; Measurement; • Software and its engineering → Data flow architectures; Cloud computing; Software performance.

show abstract

“…There is no doubt that the most fundamental problem in real-time developed applications [26,27,28] is to present the properly processed data via user-friendly interfaces. However, things may get quite complicated at this stage if architecture is not known well.…”

Section: Fundamental Difficulties In the Studymentioning

confidence: 99%

Attackdet: Combining web data parsing and real-time analysis with machine learning

Pala¹,

Şana²

2020

J. adv. tec. eng. res.

View full text Add to dashboard Cite

In this study, the web traf ic was analyzed via Machine Learning (ML) support, and incoming traf ic was visualized after real-time classi ication giving priority to stability and performance, which are indispensable for real-time applications. Websocket technology was used for instantaneous and fast data transfer. Processes may be blocked due to asynchronously operating structure when Hyper-Text Transfer Protocol (HTTP) traf ic is intensive. Synchronous operation of the system was causing both delays and negatively affecting the ef iciency of the application. To overcome this bottleneck, the developed application used asynchronous libraries instead of synchronous ones. The most essential features of the study were the analysis of HTTP packets captured in real-time, classifying the packets according to whether they are safe or suspicious using ML algorithms, and real-time display of the acquired results. In this way, incoming traf ic was classi ied smartly without getting lost in thousands of log iles. A success rate of 96.49% was attained using the logistic regression model, which is very successful in classi ication.

show abstract

Performance evaluation of real-time stream processing systems for Internet of Things applications

Cited by 30 publications

References 11 publications

Benchmarking Scalability of Stream Processing Frameworks Deployed as Event-Driven Microservices in the Cloud

Benchmarking Scalability of Stream Processing Frameworks Deployed as Event-Driven Microservices in the Cloud

How to Measure Scalability of Distributed Stream Processing Engines?

Attackdet: Combining web data parsing and real-time analysis with machine learning

Contact Info

Product

Resources

About