Out of memory errors are common challenges in database systems with vectorized query engines, particularly when handling large data fields such as JSON files in industries like cybersecurity. These engines process data in batches, often requiring substantial RAM, which can lead to errors when the memory capacity is exceeded. Traditional solutions involve adjusting block sizes, running queries on smaller data subsets, or using servers with higher RAM-to-CPU ratios. Oxla offers an innovative approach by automating batch size adjustments, ensuring that batch sizes remain within resource limits during query execution. This dynamic adjustment allows for efficient processing of large JSONs and strings, providing a more user-friendly experience without requiring developers to delve into database internals.