At 2:00 a.m each day, the database team runs a batch job to import the sales data for the day, index it, and generate sales reports for the next day. The team has informed the cloud administrator…

Rescheduling the job to start earlier only extends the available time window but does not address the underlying processing speed bottleneck, and this buffer will shrink further as data continues to grow.

Auto-scaling adds servers dynamically in response to load spikes, but batch processing jobs with fixed chunk assignments do not automatically rebalance work to newly added servers, so additional servers may sit idle.

Manually adding batch servers requires reconfiguring the job to distribute new chunks to the additional servers, introducing coordination overhead without improving the speed at which existing servers handle their current workloads.

Vertical scaling by adding CPU and memory to the existing batch servers increases raw processing throughput per server, allowing each server to complete its individual data chunk faster as data volume grows. Since the batch job already distributes work across multiple servers processing chunks individually, the bottleneck is per-server speed rather than the number of parallel workers. This directly reduces total job duration without restructuring the job distribution logic.