# Valkey benchmark **URL:** \ **Category:** Valkey & Redis\ **Created:** [March 4, 2025, 10:09pm UTC](https://forums.percona.com/t/valkey-benchmark/37042 "2025-03-04T22:09:51Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![josephtryit](https://avatars.discourse-cdn.com/v4/letter/j/49beb7/32.png) [@josephtryit](https://forums.percona.com/u/josephtryit)\ **Post date:** [March 4, 2025, 10:09pm UTC](https://forums.percona.com/t/valkey-benchmark/37042/1 "2025-03-04T22:09:51Z")
Hi, We want to compare performance between valkey and redis enterprise using valkey-benchmark for valkey and memtier\_benchmark. for redis enterprise. For redis enterprise we can connect to any of the node in the cluster and it handles the get if the hashslot is present in different node. For valkey, we get the below error when connecting to 1 out of 3 nodes. Will valkey not automatically handle if the hashslot is present in different node instead of returing the hashslot for which client has to retry? [root@xxxxxxx valkey]# valkey-benchmark -h xxxxxx -p 7000 -c 20 -P 10 --threads 2 -q PING\_INLINE: 199203.20 requests per second, p50=0.375 msec PING\_MBULK: 398406.41 requests per second, p50=0.375 msec Error from server: MOVED 13782 xxxxx:7000ll: -nan) ---
**Author:** ![martinv](https://avatars.discourse-cdn.com/v4/letter/m/e5b9ba/32.png) [@martinv](https://forums.percona.com/u/martinv)\ **Post date:** [March 5, 2025, 9:11am UTC](https://forums.percona.com/t/valkey-benchmark/37042/2 "2025-03-05T09:11:29Z")
> [@josephtryit](#): > > valkey-benchmark -h xxxxxx -p 7000 -c 20 -P 10 --threads 2 -q Hi, If you have set-up a Valkey in cluster mode, you need to run the benchmark with “–cluster” as well. Also be aware if running Valkey 8 to run enable IO threads as this can give a lot more throughput as well as lower latency. Unless you are bypassing the Redis Enterprise proxy you should get much better latency with Valkey cluster mode.