IOPS Calculator
Calculate theoretical IOPS (Input/Output Operations Per Second) and throughput based on latency and queue depth. Essential for storage performance planning.
Performance Parameters
Performance Results
Theoretical IOPS
—
Throughput
—
Read IOPS
—
Write IOPS
—
Formula:
IOPS = (1000 / Latency_ms) × Queue_Depth
Throughput_MB/s = (IOPS × Block_Size_KB) / 1024
Block Size Impact
| Block Size | IOPS | Throughput | Typical Use |
|---|---|---|---|
| 4 KB | — | — | Databases, random I/O |
| 8 KB | — | — | SQL Server, Oracle |
| 64 KB | — | — | File servers, mixed workloads |
| 256 KB | — | — | Streaming, sequential I/O |
| 1 MB | — | — | Video editing, large files |
Typical Storage Performance
| Storage Type | Typical Latency | 4K Random IOPS | Sequential MB/s |
|---|---|---|---|
| 7200 RPM HDD | 8-12 ms | 75-150 | 100-200 |
| 10K RPM HDD | 5-8 ms | 150-250 | 150-250 |
| 15K RPM HDD | 3-5 ms | 200-350 | 200-300 |
| SATA SSD | 0.1-0.5 ms | 50K-100K | 400-550 |
| NVMe SSD | 0.02-0.1 ms | 100K-1M+ | 3,000-7,000 |
| Intel Optane | 0.01 ms | 500K-2M+ | 2,500-3,000 |
Understanding Storage Performance
IOPS (Input/Output Operations Per Second) is the primary metric for measuring storage performance in random access workloads. Understanding IOPS helps you size storage systems correctly for databases, virtual machines, and transaction-heavy applications. The relationship between IOPS, latency, and queue depth follows predictable formulas that this calculator implements. Learn more: convert between storage units.
Block size significantly impacts both IOPS and throughput measurements. Small block sizes (4KB-8KB) are typical for database workloads and result in higher IOPS but lower throughput. Large block sizes (64KB-1MB) are common for sequential operations like video streaming, yielding lower IOPS but much higher throughput in MB/s. Related: Bandwidth and throughput converter.
Modern NVMe SSDs can deliver hundreds of thousands of IOPS at sub-millisecond latencies, far exceeding traditional spinning disks. When planning storage infrastructure, consider both current requirements and future growth. A system that performs well at 50% utilization may struggle as workloads increase and queue depths grow. See also: StorageMath calculators.
Key Concepts
IOPS (Input/Output Operations Per Second): Measures the number of read/write operations a storage system can perform per second.
Latency: The time delay between requesting data and receiving it. Lower is better.
Queue Depth: Number of outstanding I/O requests. Higher queue depth can increase IOPS but may also increase latency.
Block Size: The size of each I/O operation. Smaller blocks = more IOPS but less throughput. Larger blocks = fewer IOPS but more throughput.