DirectMemory Modules (DMMs) are Storage Class Memory (SCM) drives used in FlashArrays to provide a high-speed read cache (DirectMemory Cache). In the FlashArray//XL architecture, these modules provide significant performance benefits for read-intensive workloads.

A fundamental configuration rule for DMMs on the //XL platform is that they cannot be installed in a DirectFlash Shelf (DFS) .

DMMs must be installed directly into the controller chassis (the head unit) to leverage the lowest possible latency connection to the CPU and memory complex.

The external backend fabric that connects to the shelves (even though it is NVMe-oF/RoCE) adds a minute amount of latency that negates the extreme speed advantage of SCM. Therefore, Purity only supports DMMs when they are local to the controllers.

DMMs share the same physical form factor and slots as standard DirectFlash Modules (DFMs) in the chassis but are typically prioritized for specific slot ranges (e.g., the first few slots) for thermal and bus management.

=========

When an Implementation Engineer or storage administrator needs to configure the foundational networking for array-to-array replication, it is essential to distinguish between configuring the network interfaces and configuring the replication partnerships .

To set up the IP addresses, subnet masks, gateways, and assign the replication service to specific Ethernet ports (such as eth2 and eth3), the user must navigate to Settings > Network in the Purity GUI. This section of the interface is strictly dedicated to managing the physical and virtual network topology of the array. Here, the engineer defines which physical ports are allowed to handle replication traffic, ensuring it is isolated from frontend host I/O or management traffic.

It is a common misconception to navigate to "Storage > Array Connections" for this step. However, the "Array Connections" tab is used after the network settings are fully configured. "Array Connections" is where the administrator actually enters the Connection Key to establish the synchronous (ActiveCluster) or asynchronous replication link between two fully networked arrays. Therefore, the foundational IP configuration must always be completed first under the Settings > Network tab.

FlashArray performance and capacity management rely on "Write Groups"—logical groupings of drives that Purity uses to stripe data. A "Wide Write Group" typically consists of roughly 20 drives and offers better performance and capacity efficiency than smaller groups. However, Purity generally requires write groups to be contained within a single physical enclosure (chassis or shelf) to maintain failure domain boundaries and performance predictability.

In this scenario, the customer wants the new 91TB pack to form a wide write group. The chassis already has a 91TB group and a 45TB group. The shelf has one 45TB group. To create a wide write group with the new 91TB pack, it is best to pair it with an existing matching or compatible group in the same enclosure. Since the chassis is full (20 drives), you cannot simply add the new pack there without moving something.

The correct procedure is to move the 45TB write group from the chassis to the shelf . This frees up 10 slots in the chassis. The shelf now holds two 45TB groups (which is valid). The chassis now has the original 91TB group and 10 empty slots. The engineer can then install the new 91TB data pack into those empty chassis slots. Purity will then be able to merge or manage these two 91TB packs (the existing one and the new one) as a highly efficient configuration, potentially forming the requested wide write group or at least balancing the high-capacity drives within the controller chassis where bandwidth is highest. This physical rearrangement is necessary because write groups cannot span across the chassis-shelf interconnect.

=========

The FlashArray//XL170R5 is a high-performance, high-capacity enterprise storage system. To maintain its performance characteristics and data layout efficiency, Pure Storage enforces a minimum raw capacity requirement for new orders and installations.

For the //XL170R5 model, the minimum supported configuration is 115TB of raw flash. This is typically achieved through a specific population of DirectFlash Modules (e.g., twenty 5.5TB modules or a similar combination that meets the threshold).

Implementation Engineers must verify during the pre-installation site readiness review that the hardware delivered matches this minimum. Attempting to initialize an //XL170 with significantly less capacity (e.g., 90TB) would be an unsupported configuration, potentially leading to performance degradation or an inability to initialize the wide write groups required by the //XL architecture.

=========

Understanding the physical chassis layout of the Pure Storage FlashArray is essential for capacity planning and hardware installation. The FlashArray//X Gen 2 (and Gen 3) utilizes a highly efficient 3U chassis design.

When looking at the front of a standard //X chassis, there are a total of 24 physical module bays. However, these bays serve fundamentally different architectural purposes:

Capacity Drive Slots: There are exactly 20 slots strictly dedicated to capacity drives (DirectFlash Modules, or DFMs). These are numbered 0 through 19 and are where the data packs are installed to provide raw storage capacity.

NVRAM Slots: The remaining 4 slots, located centrally in the chassis, are dedicated exclusively to NVRAM modules (NVB0, NVB1, NVB2, and NVB3). These provide the highly available, non-volatile write cache for the array and cannot be used for capacity expansion.

Therefore, when an Implementation Engineer is assessing the chassis for data pack expansion or module upgrades, there are exactly 20 standard capacity drive slots available before an external DirectFlash Shelf (DFS) must be added. (Note: FlashArray//XL utilizes a larger 5U chassis with 40 capacity slots, but the standard //X remains at 20).

Performing a simultaneous data pack evacuation (capacity migration) and a Hardware Non-Disruptive Upgrade (HWNDU, such as a controller swap) introduces a high degree of complexity. Pure Storage heavily emphasizes a sequential, risk-averse methodology during these major dual-maintenance events.

The standard operating procedure dictates that these two major operations must not overlap. First, the data pack evacuation is executed. The Purity operating system transparently moves all data blocks from the targeted legacy drives to the remaining free space in the array or to a newly installed swing shelf. Once the UI and CLI explicitly show the evacuation is 100% complete and the drives are marked as "empty" or "evacuated," the Implementation Engineer can safely unlatch and physically remove those old drives.

At this precise juncture— after removing the fully evacuated drives and before starting the physical controller replacements for the NDU —the engineer must contact Pure Storage Support. Support will log into the array remotely, run pre-flight health checks (such as hardware_check.py), verify that the array has cleanly stabilized from the drive removal, and ensure the cluster is perfectly healthy before granting the "go-ahead" to initiate the first controller failover. Proceeding into an NDU without this verification risks taking a controller offline while the storage fabric is still normalizing, which could compromise data availability.

Hardware Non-Disruptive Upgrades (HWNDU) rely strictly on the array being in a fully healthy state before the process begins. The validate-array-health check is designed to gate the upgrade if any component that provides redundancy or data integrity is compromised. NVRAM (Non-Volatile Random Access Memory) modules are critical components in the FlashArray architecture; they store in-flight write data to ensure it is persistent before being destaged to flash.

A "Code 60" alert specifically points to a hardware issue with an NVRAM module. While the //X70 model indeed has redundancy (often using multiple NVRAM modules), proceeding with an upgrade while one is degraded puts the array at risk. During an NDU, controllers reboot and failover; running on reduced redundancy during these critical operations is not supported and is blocked by the health checks.

The correct first step for an Implementation Engineer is to attempt to reseat the unhealthy NVRAM module . Often, connection issues or transient seating faults can trigger these alerts. After reseating, the engineer must run the health check again to confirm the module status has returned to "healthy." If the module remains unhealthy after reseating, it must be replaced before the upgrade can proceed. You cannot simply disable it or ignore it, as the upgrade script will likely refuse to continue to protect data integrity.

=========

The pureadm command suite is the primary CLI tool used by Implementation Engineers and Support to manage the Purity operating environment's services and high-availability states. After a controller reboot—whether part of a hardware replacement, NDU, or fresh install—it is critical to verify that the Purity software stack has initialized correctly and that all sub-services (such as the I/O handling process, management daemon, and scut) are running.

The specific command pureadm list is the correct syntax to display a summary of the Purity services and their current status (e.g., running, stable, or stopped) on the local controller. This command provides a clean, immediate view of the service stack's health. If the output shows all expected services as "running," the engineer can confirm the controller has successfully rejoined the cluster and is ready to handle operations.

While pureadm alone might print help text and pureadm status is often a valid guess for other systems, the strict syntax for FlashArray implementation verification relies on pureadm list. This step is a standard part of the "Health Check" procedures found in upgrade guides. Failing to verify service status before proceeding (e.g., failing over to the other controller) could lead to an outage if the rebooted controller hasn't actually fully started its data services. Therefore, pureadm list is the validation checkpoint before declaring the controller healthy.

=========

This scenario highlights a critical architectural transition point within the Pure Storage FlashArray product line. The arrays utilize dedicated NVRAM modules to safely ingest and acknowledge incoming writes before destaging them to the capacity flash drives.

On the lower to mid-range models (such as the //X10, //X20, and //X50), the system requires only two NVRAM modules to maintain a highly available, mirrored write cache. These are physically installed in the central chassis slots NVB0 and NVB1. However, the higher-end, performance-heavy models (such as the //X70 and //X90) boast significantly more processing power and ingest bandwidth. To support these massive workloads without creating a cache bottleneck, the architecture mandates four NVRAM modules, filling slots NVB0, NVB1, NVB2, and NVB3.

Therefore, when performing an inter-generational, cross-model Hardware Non-Disruptive Upgrade (HWNDU) from an //X50 to an //X70 , the Implementation Engineer is traversing that architectural boundary. While performing the controller swaps, the engineer is strictly required to install two additional, newly shipped NVRAM modules into the previously empty bays NVB2 and NVB3. If they fail to add these modules, the new //X70 controllers will detect a cache hardware deficiency and refuse to initialize, halting the upgrade process. Options B, C, and D do not cross this 2-to-4 NVRAM threshold.

For a Non-Disruptive Upgrade (NDU) from a FlashArray//X to a FlashArray//XL , the upgrade kit typically contains 4 QSFP cables, 2 Mellanox cards, and 20 DFMD blanks (or modules) .

2 Mellanox Cards: The FlashArray//X (source) likely requires the installation of 100GbE RoCE adapters (Mellanox) into its controllers to establish the high-speed replication link with the new FlashArray//XL (which has 100GbE native). One card is installed per //X controller.

4 QSFP Cables: To ensure a redundant, high-bandwidth connection between the two arrays during the data migration phase, 2 cables are used per controller pair (CT0-to-CT0 and CT1-to-CT1), totaling 4 cables.

20 DFMDs (Blanks/Fillers): The FlashArray//X typically holds 20 DirectFlash Modules. The FlashArray//XL chassis has 40 slots. When the 20 data drives are physically moved from the //X to the //XL, the //XL will still have 20 empty slots. These must be filled with DirectFlash blanking modules (DFMD blanks) to maintain proper chassis airflow and thermal pressure. Thus, the kit includes these 20 fillers.