VM SOLUTIONS GmbH, Heidelberg, Germany: Difference between revisions
Created page with "Title: VM SOLUTIONS GmbH, Heidelberg, Germany Research Question: How can response time statistics in ERP transaction systems be better understood and measured? Methodology: The study uses heuristic rules and log-normal distribution analysis to understand response time statistics in ERP systems. Results: The study found that response time distributions in ERP systems are generally log-normal or a sum of two or more log-normal distributions. The analysis provides argume..." |
No edit summary |
||
| Line 1: | Line 1: | ||
Title: VM SOLUTIONS GmbH, Heidelberg, Germany | Title: VM SOLUTIONS GmbH, Heidelberg, Germany | ||
Research Question: How can response time statistics in ERP transaction systems | Main Research Question: How can response time statistics be improved in ERP transaction systems? | ||
Methodology: The study | Methodology: The study proposes a simple mean-field model for response time distribution in such systems. This model yields an algorithmic normal distribution of response times. The authors present data from performance measurements to support their results. The data shows that the response time distribution of a given transaction in a given system is generally a log-normal distribution or, in some situations, a sum of two or more log-normal distributions. Deviations from the log-normal form can often be traced back to performance problems in the system. | ||
Results: The | Results: The authors found that response time distributions in ERP systems are generally log-normal or a sum of two or more log-normal distributions. This suggests that performance issues in ERP systems can often be attributed to specific performance problems. | ||
Implications: The | Implications: The results have implications for the interpretation of response time data and for service level agreements. They suggest that a more detailed understanding of response time statistics in ERP systems could lead to improved performance management and potentially lower costs for enterprises. Furthermore, the findings could help in setting more effective service level agreements by providing a better understanding of what response times to expect for different transactions. | ||
Link to Article: https://arxiv.org/abs/0404035v3 | |||
Link to Article: https://arxiv.org/abs/ | |||
Authors: | Authors: | ||
arXiv ID: | arXiv ID: 0404035v3 | ||
[[Category:Computer Science]] | [[Category:Computer Science]] | ||
[[Category:Response]] | [[Category:Response]] | ||
[[Category:Time]] | [[Category:Time]] | ||
[[Category: | [[Category:Normal]] | ||
[[Category:Systems]] | [[Category:Systems]] | ||
[[Category: | [[Category:Performance]] | ||
Latest revision as of 15:48, 24 December 2023
Title: VM SOLUTIONS GmbH, Heidelberg, Germany
Main Research Question: How can response time statistics be improved in ERP transaction systems?
Methodology: The study proposes a simple mean-field model for response time distribution in such systems. This model yields an algorithmic normal distribution of response times. The authors present data from performance measurements to support their results. The data shows that the response time distribution of a given transaction in a given system is generally a log-normal distribution or, in some situations, a sum of two or more log-normal distributions. Deviations from the log-normal form can often be traced back to performance problems in the system.
Results: The authors found that response time distributions in ERP systems are generally log-normal or a sum of two or more log-normal distributions. This suggests that performance issues in ERP systems can often be attributed to specific performance problems.
Implications: The results have implications for the interpretation of response time data and for service level agreements. They suggest that a more detailed understanding of response time statistics in ERP systems could lead to improved performance management and potentially lower costs for enterprises. Furthermore, the findings could help in setting more effective service level agreements by providing a better understanding of what response times to expect for different transactions.
Link to Article: https://arxiv.org/abs/0404035v3 Authors: arXiv ID: 0404035v3