Home > Storage > PowerStore > Storage Admin > Dell PowerStore: Best Practices Guide > PowerStore deployment modes
The PowerStore platform consists of ten different models, from the PowerStore 500 model through the PowerStore 9200 model. PowerStoreOS 1.0 launched with the PowerStore 1000T/X, 3000T/X, 5000T/X, 7000T/X, and 9000T/X. PowerStoreOS 2.0 introduced the entry-level 500. PowerStoreOS 3.0 introduced a platform refresh, with the 1200T, 3200T, 5200T, and 9200T models. All models use a common base enclosure and I/O modules. The models differ by CPU core count and speed, memory size, and number of NVMe NVRAM drives. These hardware differences give each model a unique performance profile.
Besides the hardware differences between the models, PowerStore can be installed in one of three different deployment modes. Each deployment mode has different capabilities, as detailed in Table 2. Choose the deployment mode that provides the required capabilities.
External block access | External file access | AppsON functionality | |
PowerStore T model: Unified | ü | ü | X |
PowerStore T model: Block optimized | ü | X | X |
PowerStore X model | ü | X | ü |
Note: The PowerStore 500, 1200, 3200, 5200, and 9200 are only available as a T model (either unified or block optimized).
The PowerStore system has different performance characteristics depending on deployment mode.
PowerStore T models run the PowerStoreOS directly on the hardware. PowerStore T models can be installed in a unified configuration that provides file and block access, or in a block optimized configuration that provides only block access.
PowerStore T models with unified modes can provide access to block and file storage resources simultaneously. This is the default deployment mode.
If the PowerStore T model appliance will not be used for file access, it can be installed in block optimized mode, which disables the file capabilities. This mode can increase the amount of block workload that the system can provide, because it can devote the additional CPU and memory that is no longer needed for file capabilities.
PowerStore X models run the PowerStoreOS as a virtual machine on an ESXi hypervisor. This configuration allows the PowerStore X model appliance to service external host I/O and run guest VMs directly on the PowerStore hardware. PowerStore X models reserve a portion of the CPU and memory to be used for hosting user VMs. Therefore, fewer resources are available for serving external storage. The relative performance for storage from a PowerStore X model is expected to be less than that for the same PowerStore T model.
This section compares the performance potential of the different PowerStore models when serving external workload. Performance scales based on the specific hardware complement of the model and is also impacted by the configuration type.
In general, the IOPS capability of the PowerStore models scales linearly from PowerStore 500 up to 9200 models. As mentioned previously, deployment mode also impacts performance capability. A PowerStore T model in block optimized mode can deliver more block IOPS than the same model in unified mode. A PowerStore X model has less capability for block IOPS because some of the compute resources are reserved for running VMs.
Except for PowerStore 500, PowerStore systems use NVMe NVRAM drives to provide persistent storage for cached write data. PowerStore 1000 up to 3200 model arrays have two NVRAM drives per system, while PowerStore 5000 up to 9200 model arrays have four NVRAM drives per system. The extra drives mean that these systems can provide higher MBPS for large-block write workloads.
PowerStore systems can be clustered. A PowerStore cluster combines multiple PowerStore appliances into a single grouping that is managed as a single storage system. A PowerStore cluster delivers aggregate performance from all appliances in the cluster, but a single volume is serviced by only one appliance at any given time. While not required, it is recommended that all appliances in a cluster be of the same model and have similar physical capacities to provide consistent performance across the cluster. A cluster must contain only PowerStore T appliances, or only PowerStore X appliances; they cannot be mixed within a single cluster.
Volumes can be migrated between appliances in a cluster. It is recommended that any host that is connected to a PowerStore cluster has equivalent connectivity to all appliances in the cluster. All appliances in a cluster should be physically located in the same data center and must be connected to the same LAN.
Clustering is applicable to block storage resources only. While a PowerStore T model in Unified mode can serve as the cluster’s primary appliance, the file resources cannot migrate to a different appliance. When deploying multiple appliances for file access, plan to have multiple clusters.