Your SAP HANA VM (M128s) is running in Azure. During peak batch processing, HANA memory utilization reaches 95%, causing out-of-memory errors. Without changing the VM size, what is the first HANA configuration parameter to tune to reduce memory pressure?

Question

Sunny · Accepted Answer

Reduce `global_allocation_limit` and enable column store delta merge to free compressed memory. `global_allocation_limit` caps the total memory HANA can allocate (in GB). If set too high or at default, HANA may consume all available RAM. Setting it to ~90% of physical RAM leaves headroom for OS and other processes. Additionally, triggering a delta merge frees memory by compressing delta store data into the main column store. These are the first-line HANA memory tuning steps before considering a scale-up. Increasing services does not reduce memory — it adds overhead.

Your SAP HANA VM (M128s) is running in Azure. During peak batch processing, HANA memory utilization reaches 95%, causing out-of-memory errors. Without changing the VM size, what is the first HANA configuration parameter to tune to reduce memory pressure?

Your SAP HANA VM (M128s) is running in Azure. During peak batch processing, HANA memory utilization reaches 95%, causing out-of-memory errors. Without changing the VM size, what is the first HANA configuration parameter to tune to reduce memory pressure?

📚 What This Question Tests

More SAP HANA Practice Questions