Home / Benchmarks

Redis vs Memcached vs Valkey: 2026 Benchmark Data, Sourced

Memcached wins simple GET/SET at high concurrency. Redis wins batched and structured operations. Valkey 8.x wins memory efficiency vs Redis 8.x. Numbers below, sourced.

Methodology: Every number labelled “Vendor” or “Independent”. Vendor numbers come from Redis Inc, Valkey project, or a managed-service provider with a product interest. Independent numbers come from third-party engineers with no product stake. Hardware noted where the benchmark provides it. Run memtier_benchmark or redis-benchmark on your real data before committing.

Benchmark 1: Simple GET/SET throughput

Independent

Hardware: AWS c6i.2xlarge, 256-byte values  |  Source: Dev Genius, March 2026 (one independent engineer)

Memcached GET~250k ops/sec
Memcached SET~200k ops/sec
Redis GET~180k ops/sec
Redis SET~150k ops/sec
Takeaway: Memcached's multi-threaded architecture delivers a clear throughput edge on simple GET/SET at high concurrency (128+ concurrent connections): in this one test, roughly a third more operations per second. The gap widens at higher concurrency as Redis's single command thread becomes the bottleneck.

Caveat: A single independent engineer's benchmark: one hardware profile, 256-byte values, no Valkey number. Treat the magnitude as indicative, not settled; your workload may differ. Run memtier_benchmark on your own data.

Benchmark 2: Batched / pipelined operations

Independent

Hardware: AWS c6i.2xlarge, 10-op pipelines  |  Source: Dev Genius, March 2026 (one independent engineer)

Redis batched pipeline~800k ops/sec
Memcached batched~750k ops/sec
Takeaway: When your client batches requests, Redis's RESP protocol pipelining advantage narrows or reverses the Memcached throughput edge. At 10-op batches, Redis edges ahead.

Caveat: Batch size matters. Smaller batches (1-2 ops) favour Memcached; larger batches (10+) favour Redis.

Benchmark 3: Rails cache backend

Independent

Hardware: Rails 7.x application server, AWS  |  Source: Medium: Rails Cache Store Benchmarks Redis vs Memcached 2026

Memcached cache read~3ms latency
Redis cache read~12ms latency
Memcached ops/sec~60k
Redis ops/sec~50k
Takeaway: In the Rails ActiveSupport::Cache pattern, Memcached shows lower latency and higher ops/sec. This reflects Memcached's simpler protocol and multi-threaded design.

Caveat: Workload-shape dependent. Rails has a specific access pattern (small values, high concurrency). Django and Node benchmarks show different splits.

Benchmark 4: Valkey 8.0 RPS gain over Valkey 7.2

Vendor

Hardware: AWS c7g.4xlarge Graviton3, 16 vCPU  |  Source: Valkey blog: 2024 Year of Valkey

Valkey 8.01.19M RPS
Valkey 7.2~360k RPS
Takeaway: 230% RPS improvement. Valkey 8.0 redesigned I/O threading to better utilize multi-core AWS Graviton3 instances. This is a Valkey-internal comparison (8.0 vs 7.2), not a Redis vs Valkey comparison.

Caveat: Vendor-sourced. The threading redesign benefits multi-core instances most. On single-core or lower concurrency workloads, the gain is smaller.

Benchmark 5: Valkey 8.1 vs Redis 8.0 memory efficiency

Independent

Hardware: AWS c8g.2xlarge Graviton4, 8 vCPU, 50M sorted-set entries  |  Source: Momento: Valkey vs Redis Memory Efficiency at Hyperscale

Valkey 8.1 memory3.77 GB
Redis 8.0 memory4.83 GB
Takeaway: Valkey 8.1 uses ~28% less memory at 50M sorted-set entries. At hyperscale, this is a meaningful operational cost reduction. Both handle comparable RPS on this hardware (730k-1M range).

Caveat: Independent benchmark, Momento publishes Valkey-compatible managed products (Momento Cache). Neither system sustains 1M RPS on this single 8-vCPU instance. Scale hardware for production load.

Benchmark 6: Valkey multi-threaded I/O

Independent

Hardware: Multi-core instances  |  Source: Better Stack 2026 comparison

Valkey throughput improvement30-35%
Redis I/O threads (multi-threaded mode)Baseline
Takeaway: Valkey's threading redesign (8.0) delivers 30-35% throughput improvement on multi-core instances compared to Redis's optional I/O threading model. Advantage grows with core count.

Caveat: Multi-core instance dependent. On single-core instances, Valkey's threading provides little advantage.

Benchmark 7: Redis 8.0 latency improvements

Vendor

Hardware: Vendor hardware (not disclosed)  |  Source: Redis blog: Redis 8 GA

Bitmap latency reductionUp to 78%
SET command latency reductionUp to 66%
Sorted set latency reductionUp to 58%
Takeaway: Redis 8.0 claims significant latency improvements vs Redis 7.x. Highly workload-specific. Replicate independently with your data before betting on these numbers.

Caveat: Vendor-sourced. Redis 8.0 vs Redis 7.x comparison only (not vs Valkey or Memcached). Hardware not disclosed. 'Up to' figures represent peak conditions.

Synthesis table

WorkloadWinnerSource typeApprox delta
Simple GET/SET, high concurrencyMemcachedIndependent~30-40% (one test)
Batched 10-op pipelineRedisIndependent~7%
Rails cache layerMemcachedIndependent~20% ops/sec, lower latency
Memory at 50M sorted-set entriesValkey 8.1Independent~28% lower
Multi-threaded I/O (multi-core)ValkeyIndependent~30-35%
Valkey 8.0 vs 7.2 RPSValkey 8.0Vendor230% (vendor claim)
Redis 8.0 vs 7.x latencyRedis 8.0VendorUp to 78% (vendor claim)

What the numbers do not tell you

Every benchmark is workload-shaped. A 256-byte simple GET/SET benchmark on a c6i.2xlarge says nothing about your 4KB session blobs, your sorted-set leaderboard under 50k concurrent users, or your pub/sub fan-out to 200 subscribers. The performance delta between Redis and Memcached is real but smaller than the license delta for many 2026 teams. Run your own benchmark with memtier_benchmark before deciding.

License timeline →Architecture explanation →Valkey vs Redis in depth →

Didn't find your answer?

Ask us. A real person reads every question and we answer the ones we can, with sources. If your question would help other readers, we may publish an anonymised version, with your permission. General reference only.

Ask a question

questions@redisvsmemcached.com