Elements for Response Time Statistics in ERP Transaction Systems
Andreas Mielke
Abstract
We present some measurements and ideas for response time statistics in ERP systems. It is shown that the response time distribution of a given transaction in a given system is generically a log-normal distribution or, in some situations, a sum of two or more log-normal distributions. We present some arguments for this form of the distribution based on heuristic rules for response times, and we show data from performance measurements in actual systems to support the log-normal form. Deviations of the log-normal form can often be traced back to performance problems in the system. Consequences for the interpretation of response time data and for service level agreements are discussed.
Create a lesson
Related papers
SysPilot: Cross-Library Performance Autotuning via Library Performance Contracts
Yineng Yan, Haowei Chen, Christopher J. Rossbach
Memory-Access Profiling for Early Multicore Interference Awareness in Embedded Systems
Jean-Baptiste Lefoul, Ahmad Shahnejat Bushehri, Alaric Chan Lock et al.
Easy Multi-Precision Acceleration via Vectorization
Chansup Byun, Piotr Luszczek, Jeremy Kepner et al.
Software Design Patterns for Latency Measurements with eBPF in SDN Environments
Pedro Lucas M. S. Rosa, Bruno L. Dalmazo, Leticia S. Machado et al.
Trustworthiness as a Measurement Structure for Task Conformance in 6G Service Chains
Bin Han, Xintong Ling, Muxia Sun et al.
Warpline: Binary-Level Tracing and Unified Evidence for Agentic GPU Kernel Optimization
Junyu Yin, Vinod Grover, Keren Zhou