Aegis Managed Pure Storage: Co-Managed FlashArray Operations

Pure Storage delivers one of the operationally simplest enterprise storage platforms available. Evergreen provides non-disruptive controller upgrades. Pure1 delivers proactive health telemetry. Purity OS upgrades execute without downtime. Drive failures get handled proactively.

The question becomes: what does co-managed Pure actually cover?

The answer is operational attention and coordination, not complexity management.

Pure environments still require human judgment for operational continuity. Purity OS upgrades need scheduling around business requirements, testing against application workloads, and execution within coordinated change windows. Pure1 alerts require assessment, prioritization, and action. Capacity management stays proactive — Pure alerts on utilization thresholds, but someone must act before capacity exhaustion.

Volume and host management requires changes when compute nodes change — new hosts added, nodes decommissioned, workloads moved. Performance anomalies like noisy-neighbor VMs or unexpected IOPS spikes need per-volume investigation that crosses Pure's boundary into hypervisor territory. NVMe-oF path health spans Pure initiator sessions and the network fabric Ethernet path — neither vendor's monitoring spans the boundary.

Aegis Managed Pure Storage provides purpose-built monitoring, incident response, and lifecycle governance for Pure FlashArray environments. The service integrates with Pure's existing Evergreen support model rather than replacing it, focusing on operational coordination and cross-platform visibility that Pure's native tools don't provide.

The three core components — monitoring via Aegis PM, incident response via Aegis IR, and lifecycle management via Aegis LM — work together to provide comprehensive operational coverage while leveraging Pure's proactive hardware management through Evergreen.

LogicMonitor's native Pure FlashArray datasource polls via Pure REST API v2, providing comprehensive visibility across array performance, capacity, health, and replication status.

Array-level performance monitoring covers read and write IOPS, latency measured in microseconds with alert thresholds calibrated to workload baseline rather than generic percentiles, throughput in GB/s, and queue depth for pending I/O operations. Sustained queue depth indicates storage pressure before latency spikes occur.

Capacity monitoring tracks array used capacity in both raw and effective terms, provisioned capacity versus physical capacity to monitor overprovision ratios, and data reduction ratios combining deduplication, compression, and thin provisioning. Reduction ratio degradation alerts when data mix changes. Per-volume capacity monitoring identifies volumes consuming disproportionate capacity, while snapshot space consumption tracking prevents snapshot accumulation from consuming capacity without volume-level alerting.

Health monitoring covers array overall health status, individual drive health including wear indicators, controller health tracking active versus passive states and failover conditions, and environmental factors like fans, temperature, and power supplies. Pure FlashArray NVRAM health monitoring is critical for write performance.

Replication monitoring tracks replication session status per protection group, replication lag, and remote array connectivity.

Per-volume IOPS monitoring identifies noisy-neighbor VMs consuming disproportionate storage IOPS. When correlated with hypervisor VM data, IVI maps high-IOPS volumes to specific VMs for investigation. This cross-platform correlation requires custom workflow in Aegis PM, not native LogicMonitor functionality.

Custom monitoring tracks active NVMe-oF initiator sessions from Pure's perspective, correlates with network interface metrics for end-to-end path health, and alerts when session count drops below expected host count. This indicates path failure before storage errors appear in VMs.

Aegis IR provides IVI first call response for array performance degradation with diagnosis across Pure storage layer, network fabric path, and hypervisor host/VM layer for full stack investigation. Replication failures get diagnosed and remediated including connectivity issues and reseeding when required.

Capacity alerts trigger client notification, growth trajectory assessment, and coordinated response. Controller failover events require confirmation that secondary controller is active, Pure1 alert acknowledgment, and beginning Evergreen coordination for hardware events. Drive failure events typically trigger Pure1 alerts and Evergreen replacement coordination — IVI confirms, tracks replacement tickets, and validates post-replacement cluster rebuild completion.

Volume management changes including host add/remove operations, volume group changes, and snapshot schedule management fall under Aegis IR scope.

Guest OS I/O issues inside VMs that manifest as storage performance problems remain client responsibility — IVI proves storage layer health but what runs above it stays with the client. Application-level data management including database backups and snapshot policies driven by application requirements are outside scope, though Pure snapshot schedules are managed. Pure1 license management and contract renewals are not included.

Purity OS lifecycle management covers tracking Pure's regular release cadence for both major releases and maintenance releases. IVI tracks recommended upgrade versions via Pure's release notes and recommendation channels, plans quarterly upgrade schedules or follows Pure recommendations, coordinates change windows with clients, and executes non-disruptive upgrade procedures with post-upgrade performance and health validation.

Pre-upgrade procedures include reviewing Pure1 health, confirming no open critical alerts, validating NVMe-oF path health, and confirming host connectivity.

Hardware EOL and Evergreen planning leverages Pure's Evergreen model which includes controller and hardware technology refreshes as part of subscription. IVI tracks when refreshes are scheduled and coordinates planning around the environment. Pure1 provides controller upgrade recommendations that IVI acts upon.

CVE tracking monitors Pure Security Advisories and PSIRT notices with defined patching timelines.

When a workload experiences performance problems, the symptom often appears in the application as slow response time, but the cause could be in the Pure FlashArray with high queue depth, latency spikes, or volumes hitting limits. It could be NVMe-oF path issues like interface errors, path failures, or MTU mismatches. The hypervisor might have CPU saturation, scheduling delays, or live migration load. Or the guest OS could have applications consuming excessive I/O or runaway processes.

None of these four monitoring domains — Pure, network fabric, hypervisor, or guest OS — has native visibility into the others. Without Aegis, teams manually investigate each layer, often spending hours on the wrong layer.

Aegis PM's cross-platform correlation pulls from Pure via LogicMonitor, from network fabric via API, from hypervisor via management API, and from compute via management interface to isolate problems to the correct layer quickly. Per-volume IOPS from Pure combined with VM-to-volume mapping from the hypervisor identifies a noisy VM. Network interface error correlation with Pure session counts identifies a path failure.

Aegis Managed Pure fits organizations running disaggregated architecture with hypervisor plus Pure FlashArray who want unified operations management across both platforms through a single Aegis engagement. It serves organizations with Pure FlashArray who want operational monitoring, change coordination, and lifecycle management handled without building internal Pure expertise.

Organizations that find Purity OS upgrade coordination and Evergreen management more time-consuming than expected benefit from the service. Those with compliance requirements where storage change documentation, firmware currency, and incident records require audit trails find value in the structured approach.

Evaluating co-managed Pure operations requires answering four key questions that determine fit and scope.

First, are you running Pure FlashArray as external storage for a hypervisor in a disaggregated architecture? If yes, Aegis Managed Pure extends other Aegis managed services naturally through a single managed services engagement covering multiple platforms.

Second, do you have internal expertise to manage Purity OS upgrades and coordinate Evergreen hardware events? If yes, consider a monitoring-only engagement with Aegis PM for Pure while the client handles lifecycle events. If no, full Aegis scope including lifecycle management for Pure provides more value.

Third, is cross-platform performance correlation between Pure and hypervisor a current pain point? If yes, this represents Aegis PM's core value for disaggregated environments. If no, it may be lower priority but remains relevant for future incident prevention.

Fourth, are you under Evergreen support with Pure? If yes, IVI coordinates Evergreen events effectively. If not under Evergreen, evaluate whether support contract renewal should be part of the engagement scope, as IVI should advise on Pure support options as part of planning.