Redis INFO Memory Explained: Understanding Redis Memory Statistics

Published: 2026-07-17

Introduction

Redis stores its dataset in memory, making memory usage one of the most important metrics to monitor in production environments.

The INFO memory command provides detailed information about Redis memory allocation, fragmentation, peak usage, and configured memory limits. Understanding these statistics helps administrators troubleshoot memory issues, optimize performance, and prevent unexpected out-of-memory conditions.

This guide explains the most important fields returned by INFO memory and how to interpret them.

What Is INFO Memory?

The INFO memory command returns memory-related statistics about a running Redis server.

Run:

redis-cli INFO memory

Example output:

Memory

used_memory:1528344

used_memory_human:1.46M

used_memory_rss:6291456

used_memory_peak:3145728

used_memory_peak_human:3.00M

maxmemory:0

mem_fragmentation_ratio:4.12

used_memory

Represents the amount of memory currently used by Redis.

This includes:

Dataset memory

Internal data structures

Redis overhead

Example:

used_memory:52428800

means Redis is currently using approximately 50 MB of memory.

used_memory_human

Displays the same value in a human-readable format.

Example:

used_memory_human:50.00M

This field is intended for administrators and monitoring dashboards.

used_memory_peak

Shows the highest amount of memory Redis has used since startup.

This value is useful for:

Capacity planning

Detecting memory spikes

Performance analysis

used_memory_rss

Represents the physical memory allocated by the operating system.

It is often larger than used_memory because of:

Memory allocator overhead

Fragmentation

Reserved memory

maxmemory

Displays the configured memory limit.

Example:

maxmemory:1073741824

This indicates a limit of 1 GB.

If the value is:

maxmemory:0

Redis has no configured memory limit.

For more information, see:

Redis maxmemory Explained

mem_fragmentation_ratio

Indicates the relationship between physical memory and Redis memory usage.

Formula:

used_memory_rss / used_memory

Typical values:

| Ratio      | Meaning                |
| ---------- | ---------------------- |
| Around 1.0 | Excellent              |
| 1.1–1.5    | Normal                 |
| Above 1.5  | Possible fragmentation |
| Above 2.0  | High fragmentation     |

High fragmentation may indicate that restarting Redis could reclaim memory.

allocator_frag_ratio

Measures fragmentation reported by the memory allocator.

A consistently high value may indicate inefficient memory allocation patterns.

allocator_rss_ratio

Shows the difference between allocator memory and operating system memory usage.

This metric is useful when investigating unexpectedly high RSS memory consumption.

Monitoring Memory Usage

Run periodically:

redis-cli INFO memory

or monitor automatically using:

Prometheus

Grafana

Redis Insight

Tracking memory trends over time is often more valuable than examining a single snapshot.

Common Memory Problems

No Memory Limit

maxmemory:0

Without a configured limit, Redis may consume all available system memory.

High Fragmentation

Example:

mem_fragmentation_ratio:3.2

Possible causes include:

Frequent key deletion

Large temporary datasets

Long-running Redis instances

Memory Reaching the Limit

When used_memory approaches maxmemory, Redis begins applying the configured eviction policy.

See:

Redis maxmemory-policy Explained

Best Practices

Monitor used_memory continuously.

Configure an appropriate maxmemory value.

Review fragmentation ratios regularly.

Investigate unexpected memory growth.

Monitor peak memory usage for capacity planning.

Combine INFO memory with other monitoring tools for long-term analysis.

Redis Explained

Redis maxmemory Explained

Redis maxmemory-policy Explained

Redis Persistence Explained

Redis INFO Explained (coming soon)

Conclusion

The INFO memory command provides valuable insight into how Redis uses memory and is one of the most important tools for monitoring production servers. By understanding metrics such as used_memory, maxmemory, used_memory_peak, and mem_fragmentation_ratio, administrators can detect memory issues early, optimize performance, and plan system capacity more effectively.

Regular monitoring of these statistics helps ensure that Redis remains stable, efficient, and ready to handle increasing workloads.

💡 Webmaster Recommendation: Facing network latency or connection drops across Southeast Asia? Try our Global Network Accelerator & Proxy Recommendation to optimize your browsing metrics.