Performance Predictions for Data Streaming Applications with Component Models

Karlsruhe Institute of Technology researchers published this dataset supporting their article at the 16th European Conference on Software Architecture. It contains measurements, models, simulation results, and analysis code for evaluating data streaming application performance. The work was supported by KASTEL Security Research Labs and the German Research Foundation.

Use Cases

• Validating performance prediction models based on the provided system measurements and simulation results.
• Benchmarking simulation tools for data streaming applications using the included models and code.
• Analyzing the accuracy of component-based performance models for streaming architectures.
• Reproducing the research evaluation for the associated academic paper.

Strengths

• Includes multiple data artifacts: measurements, models, simulation results, and analysis code.
• Supports a peer-reviewed publication from a major software architecture conference (ECSA).
• Associated with a recognized research institution (Karlsruhe Institute of Technology) and funded projects.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and file size are unknown, which may limit suitability assessment.

Provenance

Source

Dominik Werle, Stephan Seifermann, Anne Koziolek (Karlsruhe Institute of Technology)

Collection Method

Likely gathered from experimental measurements and simulations of a data streaming system.