PowerFlex supports traditional disaggregated (two-layer) infrastructure architecture, single-layer HCI architecture, or a mix of both configurations.
In a single-layer HCI deployment (shown in the figure below), the SDCs and SDSs run on the same set of nodes. This deployment maximizes hardware utilization and reduces the infrastructure footprint. This approach works very well in environments where a single group manages both storage and compute. In this architecture, each node in the cluster contributes to storage resources as well as hosts applications and workloads.
In a PowerFlex disaggregated (two-layer) deployment (shown in the figure below), compute resources exist on one set of nodes and storage resources exist on another set of nodes. Disaggregated deployments also allow compute and storage resources to grow independently. In disaggregated deployments, existing compute and storage teams can work together to build a large-scale, software-defined infrastructure. This segregation of compute and storage resources can be helpful to minimize software licensing costs in certain situations.
PowerFlex offers the flexibility for users to mix the previous two deployment approaches to create a configuration that best meets their needs. Using PowerFlex compute-only nodes, enable customers to scale compute resources. Using PowerFlex storage-only nodes, enable customers to scale storage resources. This type of architecture delivers ultimate flexibility. The figure below shows a mixed architecture deployment.