Automated Debugging in a Trading System
Verifying the reliability and functionality of a complex system like a trading system is highly demanding since failure in such a system can cause serious economic problems. Automated random testing is a good solution to find new and rare failures in such a system. Test cases in random testing usual...
| Main Author: | |
|---|---|
| Format: | Others |
| Language: | English |
| Published: |
KTH, Skolan för informations- och kommunikationsteknik (ICT)
2012
|
| Subjects: | |
| Online Access: | http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-90331 |
| Summary: | Verifying the reliability and functionality of a complex system like a trading system is highly demanding since failure in such a system can cause serious economic problems. Automated random testing is a good solution to find new and rare failures in such a system. Test cases in random testing usually contain a long sequence of actions that debugging them manually to find the root cause of the failure is a very boring and tiresome task. This thesis aims to create a model for automating the task of the debugging to reduce the failed test case to an equivalent test case that only contains relevant actions that together cause the failure. Delta debugging is the core algorithm of the model that simplifies a failed test case by successive testing. The target of the project is TRADExpress system of Cinnober Financial Technology AB. The model is integrated to the random testing framework of the TRADExpress system. |
|---|