How can you optimize cost when running Fabric workloads?

• A. Increase compute resources across the board to guarantee performance.
• B. Right-size compute, select appropriate pool sizes, scale to demand, and leverage caching and materialized views; monitor usage with cost dashboards.
• C. Disable caching to reduce memory usage.
• D. Use a single fixed pool for all workloads regardless of demand.

Answer: (B)

Optimizing cost with Fabric workloads centers on sizing resources to actual needs and using architectural techniques that reduce unnecessary compute while preserving performance. Right-sizing compute means choosing pool sizes that match the workload, avoiding over-provisioning, and using autoscaling to grow and shrink resources as demand changes. Scaling to demand ensures you allocate more power when workloads spike and release it when they lull, which prevents paying for idle capacity. Leveraging caching and materialized views helps because frequently accessed data can be served from fast, precomputed results rather than re-running expensive queries, dramatically cutting compute cycles and latency. Caching improves responsiveness and materialized views precompute results for common queries, both lowering ongoing costs for heavy workloads.

Monitoring usage with cost dashboards ties it all together by giving visibility into where spend is going, so you can spot waste, adjust pool configurations, and set budgets or alerts. In contrast, increasing compute resources across the board without regard to actual need leads to unnecessary expenses, disabling caching can increase compute and data-processing costs, and using a single fixed pool eliminates elasticity, forcing either wasted capacity or performance bottlenecks.