Track 6: Performance Monitoring, Modeling, Analysis, and Benchmarking
Performance is a key challenge in cloud, cluster, and Internet computing infrastructure. Topics of interest include, but are not limited to:
- Performance metrics and models: Performance in broad sense, performance, elasticity, efficiency, wear. Model construction and validation, modeling formalisms, model inference. Models in the loop.
- Analysis of system or application performance: Exploration techniques, performance testing, statistical and machine learning approaches. Anomaly detection. Visualization.
- Monitoring and evaluation tools: Introspection techniques, accuracy and overhead, continuous data collection, profiling. Cloud-native performance tools.
- Performance management: Resource accounting and scheduling. Automated data and code placement. Handling performance anomalies, system resiliency and adaptivity.
- Performance benchmarking: Workload characterization. Black box platform benchmarking. Scalability. Reproducibility. Research and evaluation datasets.
- Systems performance: Performance in cloud, edge and fog systems. Serverless performance. Emerging systems for machine learning, AI, and quantum computing. Disaggregated storage and memory. Accelerators.
Hiroyuki Takizawa, Tohoku University, Japan
Petr Tuma, Charles University, Czech Republic
Janki Bhimani, Florida International University, USA