Performance and scalability

Scalability and performance

Netezza Performance Server – Software is designed with a modular and scalable architecture, enabling customers to independently scale compute and storage resources based on workload demands, data volume, and performance goals.

Compute scalability

Compute resources can be scaled by adding more nodes to the OpenShift cluster. This allows additional SPUs (Snippet Processing Units) to be scheduled, enhancing parallel processing and increasing query throughput.

This approach supports growing workloads and concurrent users without requiring architectural changes.

Storage scalability

Storage is managed by OpenShift Data Foundation (ODF), which allows scaling by adding new storage devices or expanding existing storage pools. Customers can build storage-dense systems by attaching larger or additional NVMe/SSD volumes to storage nodes.

The underlying Ceph cluster in ODF automatically adjusts to added capacity, enabling seamless storage scaling without downtime.

Independent scaling of compute and storage

The compute and storage layers are decoupled, allowing customers to scale each independently:

• Add compute nodes to enhance performance
• Add storage nodes or volumes to increase capacity
• Scale both based on specific business or technical requirements

Performance optimization

Performance can be further optimized by:

• Adding more SPU nodes to support query-intensive or high-concurrency environments
• Using high-speed NVMe drives for transient storage (nzlocal) to accelerate specific operations