Home / Benchmarks

Redis vs Memcached vs Valkey: 2026 Benchmark Data, Sourced

Memcached wins simple key-value caching at high concurrency. Redis wins 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 key-value caching under production load

Independent

Hardware: 8-core AWS instances (type not named), ~300-byte objects, 150,000 req/s peak, 80% reads  |  Source: Dev Genius, 12 March 2026 (one team's production test)

Memcached average latency0.5 ms
Redis average latency0.8 ms
Memcached P99 latency1.1 ms
Redis P99 latency3.4 ms
Memcached memory, 40M keys8 GB
Redis memory, 40M keys12 GB
Takeaway: On simple key-value caching (product availability, sessions, rate-limit counters) Memcached posted lower average latency, about a third of Redis's P99 and a third less memory. The author traces it to Redis's single command thread maxing out one core at 150,000 requests per second, while Memcached spread the load across all 8 cores.

Caveat: One team's production test, published as a blog post. The author does not name the instance type and reports no ops/sec ceiling. Treat it as one data point, not a settled figure; run memtier_benchmark on your own data.

Benchmark 2: Structured data (shopping carts)

Independent

Hardware: Same 8-core AWS setup; carts held as lists, recently viewed as sorted sets  |  Source: Dev Genius, 12 March 2026 (same production test)

Redis (LPUSH, LREM, ZADD)~5x faster (author's figure)
Memcached (serialise whole cart)Baseline
Takeaway: Where the data has structure, Redis wins clearly. Memcached has to read, deserialise, change and rewrite the whole cart on every update; Redis changes one element in place. The same team kept Memcached for about 80% of its caching and Redis for the 20% that needed lists, sorted sets and pub/sub.

Caveat: The author states the 5x figure without publishing the raw numbers behind it. The direction (native data structures beat whole-object rewrites) is the durable lesson; the multiple will vary with object size.

Benchmark 3: Valkey 8.0 RPS gain over Valkey 7.2

Vendor

Hardware: AWS EC2 c7g.16xlarge, 8 I/O threads, 512-byte values, 650 clients, SET  |  Source: Valkey blog: Unlock 1 Million RPS

Valkey 8.01.19M RPS
Valkey 7.2360k RPS
Takeaway: About 230% more throughput, with average latency down 69.8% (1.792 ms to 0.542 ms). The gain comes from the redesigned I/O threading plus memory prefetching. This is a Valkey-internal comparison (8.0 vs 7.2), not a Redis vs Valkey comparison.

Caveat: Vendor-sourced, on a 64-vCPU instance with 8 I/O threads. On small instances or low concurrency the gain is smaller.

Benchmark 4: Valkey 8.1 vs Redis 8.2 memory efficiency

Vendor

Hardware: AWS c8g.2xlarge Graviton4, 8 vCPU, 16 GB, 50M sorted-set entries  |  Source: Momento: Valkey vs Redis Memory Efficiency at Hyperscale

Valkey 8.1 memory3.77 GB
Redis 8.2 memory4.83 GB
Takeaway: Valkey 8.1 used 1.06 GB (28%) less memory for 50 million sorted-set entries. Insert throughput finished at about 570k/s for Valkey and 530k/s for Redis, a ~7% gap.

Caveat: Momento sells Valkey-compatible managed services, so it has a product interest; the benchmark code is published for anyone to rerun. Sorted sets only; other data types will differ.

Benchmark 5: Redis 8.0 vs Redis 7.2.5

Vendor

Hardware: Vendor hardware; io-threads test on a multi-core Intel CPU  |  Source: Redis blog: Redis 8 is now GA

p50 command latency reduction5.4% to 87.4%
Throughput with io-threads=8Up to 112% more
Takeaway: Redis reports 90 of 149 command tests ran faster on Redis 8 than on 7.2.5, with p50 latency cuts from 5.4% to 87.4%. Its new I/O threading (off by default; set io-threads) measured up to 112% more throughput at 8 threads.

Caveat: Vendor-sourced, Redis 8 vs Redis 7.2.5 only (not vs Valkey or Memcached). 'Up to' figures are best cases; the median command gains far less.

Synthesis table

WorkloadWinnerSource typeApprox delta
Simple key-value, production loadMemcachedIndependentP99 1.1 vs 3.4 ms; 8 vs 12 GB (one team)
Structured data (carts: lists, sorted sets)RedisIndependent~5x (author's figure)
Memory at 50M sorted-set entriesValkey 8.1Vendor28% lower vs Redis 8.2 (Momento)
Valkey 8.0 vs 7.2 RPSValkey 8.0Vendor~230% (vendor claim)
Redis 8.0 vs 7.2.5 p50 latencyRedis 8.0VendorUp to 87.4% (vendor claim)
Redis 8.0 io-threads=8 throughputRedis 8.0VendorUp to 112% (vendor claim)

What the numbers do not tell you

Every benchmark is workload-shaped. A 300-byte key-value test on 8-core instances 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