What Metrics Should You Monitor in ftScalable Storage Systems?

For organizations running mission-critical workloads on Stratus ftServer environments, the ftScalable storage array is a backbone component, purpose-built for fault tolerance, high availability, and scalable capacity. But even the most reliable hardware requires active monitoring. Knowing which metrics to track, and when to act on them, is the difference between proactive storage management and an unplanned outage.

This post walks storage administrators and infrastructure leads through the key performance indicators to watch in an ftScalable environment and how Sightline EDM makes monitoring those indicators actionable.

Q: What is ftScalable storage, and why does it require specialized monitoring?

The Stratus (Pengiun Solutions) ftScalable storage array is a next-generation hybrid storage solution designed to complement ftServer systems. It supports RAID levels 1, 5, 6, 10, and ADAPT across up to 248 drive slots and can deliver up to 600,000 IOPS, making it well-suited for high-performance, always-on application environments. Its architecture includes real-time data tiering between performance, standard and archive tiers, rapid RAID rebuild (performing 2–5x faster than standard rebuild designs), and support for both SSD and HDD configurations.

Because ftScalable environments often underpin critical applications in financial services, manufacturing, utilities, and government operations, even small degradations in storage performance can have outsized consequences. Specialized monitoring matters because the system’s complexity, spanning multiple RAID groups, tiered storage, and high IOPS workloads, produces a wide range of signals that standard server monitoring tools may not capture with the required granularity.

Q: What are the most important IOPS and throughput metrics to track?

IOPS (Input/Output Operations Per Second) and throughput are the primary indicators of storage performance. In an ftScalable environment, the metrics worth tracking most closely include:

• Disk I/Os per second – at both the individual disk and volume level
• Read/write IOPS split – understanding the read-vs-write ratio reveals workload patterns and helps size cache appropriately
• KBytes transferred per second – at the host port, virtual disk, and controller levels
• Write cache hits/misses and read cache hits/misses – cache efficiency is a leading indicator of overall I/O health
• Queue depth – sustained high queue depths at the disk or controller level often signal saturation before latency spikes become visible

Sightline collects all of these metrics in real time at the disk unit, volume, host port, virtual disk, and controller levels, giving storage admins a complete, unified view of I/O activity across the entire ftScalable array.

Q: How should I be monitoring disk latency in ftScalable environments?

Disk latency is one of the most direct measures of storage health and user experience. In ftScalable systems, the relevant latency metrics include:

• Average response time – the total time from I/O request to completion
• Average service time – the disk-level processing time, isolated from queue delays
• Average queue time – how long I/Os wait before being serviced; sustained queue times are a leading indicator of saturation
• Read vs. write response time – separated read and write latency metrics allow for more precise root cause identification

Trending these metrics over time, not just monitoring point-in-time values, is essential. A latency number that looks acceptable in isolation can be a warning sign when it represents a consistent upward trend. Sightline EDM’s over-time data storage and historical analysis capabilities allow teams to distinguish transient spikes from genuine degradation trends.

Q: What RAID health indicators should storage admins watch?

RAID health is where proactive storage monitoring has the highest stakes. ftScalable systems support RAID 1, 5, 6, and 10, each with different rebuild behaviors and fault tolerance thresholds. The critical indicators to track include:

• Disk degradation status – one of the earliest signals that a disk is underperforming and may be approaching failure
• P1/P2 SMART count metrics – including I/O timeout count, no-response count, and spinup retry count; these SMART indicators often predict failure days before it occurs
• Media errors and non-media errors – sudden increases in either category warrant immediate attention
• Block reassigns and bad blocks – accumulation of bad blocks accelerates ahead of disk failure in spinning media
• Fatal errors and data errors per second – even low rates require investigation in a fault-tolerant environment
• Recovery switches per second – an elevated rate indicates the system is frequently entering a recovery state to reconcile data, often a precursor to a total disk group failure

ftScalable’s rapid rebuild capability is an architectural advantage but administrators still need to know when a rebuild is occurring, how long it is taking, and whether the RAID group is operating in a degraded state during that period. Monitoring tools that surface these conditions in real time are essential for maintaining the five-nines availability target these systems are designed to deliver.

Q: How does monitoring tie into redundancy planning?

Redundancy planning isn’t a one-time activity, it’s an ongoing operational discipline. Historical performance data from your ftScalable environment feeds directly into capacity planning decisions: when to add shelves, when to migrate workloads off aging spindles, when SSD read-cache is delivering returns, and when a RAID group’s workload profile has drifted from its original design assumptions.

Sightline EDM’s ForSight® predictive analytics and capacity planning capabilities allow infrastructure teams to forecast storage growth, model future demand, and make evidence-based decisions about expansion or reconfiguration before capacity constraints create performance problems.

Q: How do Sightline dashboards and alerts work for ftScalable monitoring?

Sightline EDM is a certified partner solution for Stratus ftScalable storage and collects granular, real-time performance data across all layers of the array: disk units, volumes, host ports, virtual disks, and controllers. Key capabilities for storage admins include:

• Custom dashboards with configurable views of storage performance data, organized by workload, RAID group, or facility
• Threshold-based alerting across any tracked metric, allowing teams to get ahead of issues before they impact applications or users
• Clairvor® correlation and root-cause analysis, which connects storage performance events to related infrastructure metrics for faster forensic investigation
• Centralized data storage that supports historical trend analysis and over-time reporting across multi-vendor environments—including SMI-S compliant storage from other vendors
• Enterprise-wide visibility across cross-platform, multi-component infrastructure in a single pane of glass

Sightline EDM is sold and supported by Stratus as an authorized partner solution for ftScalable storage environments. For a complete list of disk statistics, volume statistics, and controller metrics captured by Sightline for ftScalable, visit stratus.com or contact your Stratus representative.

The Bottom Line

ftScalable storage is engineered for continuous availability, but availability is maintained, not assumed. Storage admins and infrastructure leads who actively monitor IOPS, disk latency, RAID health indicators, and capacity trends are the ones who keep that availability promise intact. By moving from “watching for breaks” to “monitoring for health,” you ensure that the only time people notice the storage is when you’re showing off your uptime reports.

Don’t wait for a bottleneck to become the headline. Sightline EDM provides the real-time visibility, historical depth, and alerting intelligence to make that monitoring practical and actionable—whether you’re managing a single ftScalable array or an enterprise-wide, multi-vendor storage environment. Contact us today to learn how Sightline can strengthen your ftScalable storage monitoring strategy.