HotKV Cloud vs Google Cloud Memorystore

A Memorystore alternative on the same cloud

HotKV Cloud is a Memorystore alternative on the same cloud. It runs on Google Cloud in 42 regions, with a multi-threaded engine that speaks the Redis protocol on a dedicated machine per instance. Memorystore runs Valkey, Redis Cluster, Redis or Memcached. Same cloud, same clients, a different engine.

Get early access See the measurements

Same region. Same clients. A different shape of server.

Memorystore

a shard per node, one thread executes

HotKV Cloud

one isolated machine, every core executes

An illustration, not a measurement. Both run on Google Cloud, and HotKV Cloud offers 42 regions. The same clients connect to either. Memorystore for Valkey and for Redis Cluster spreads keys across nodes by hash slot, and each node executes commands on one thread. A HotKV Cloud instance is one isolated machine whose engine executes on every core.

This Memorystore alternative, run as a single instance on one 160-vCPU machine, led 12 of 12 commands against Valkey 9.1.2, at 8.6x to 56.6x the throughput. HotKV was built from source for that CPU. Valkey 9.1.2 ran as its stock build. The gap is smaller on small machines.

Drop-in

What stays the same with this Memorystore alternative

HotKV implements the Redis wire protocol. Moving from Google Cloud Memorystore is usually a change of host and credentials. Check the command reference for anything outside the core set.

Same protocol

Redis-compatible. RESP2 and RESP3 with HELLO negotiation, and HotKV reports Redis version 8.4.0 to clients, so libraries pick their features by it. Connection strings, pipelines, pub/sub and transactions work as they do today.

Same clients and tools

Redis and Valkey client libraries, redis-cli, redis-cli --cluster and GUI tools connect to HotKV without code changes for the core commands.

Same core commands

300 core commands, Lua scripting and function libraries, plus HotKV's own command families when you want them.

the client you already use
$ redis-cli -h <your-hotkv-host> -p 6379 --tls --user app --askpass
> HELLO 3
1# "server" => "hotkv"
2# "version" => "8.4.0"
3# "proto" => (integer) 3
   ... (four more fields)
> SET session:4821 "same protocol, every core"
OK
> GET session:4821
"same protocol, every core"
> TAG.SET session:4821 tenant:acme   # a HotKV command, when you want it
(integer) 1

The case for a Memorystore alternative

Why teams look for a Memorystore alternative

Documented properties of Google Cloud Memorystore that send teams looking for a Memorystore alternative, checked on October 21, 2026.

01

The engines execute commands on one thread

Valkey and Redis execute commands on a single main thread per node. Memorystore scales by adding nodes and shards, which you size and pay for. A Memorystore alternative with a multi-threaded engine scales inside one machine first.

02

Shards and nodes to plan

Memorystore for Valkey and for Redis Cluster scale horizontally up to 250 nodes. Capacity is planned in shards and replicas, and multi-key commands must stay within a hash slot. A Memorystore alternative on one machine has no slots to plan.

03

AI workloads need more services

A prompt cache, semantic cache, agent memory and LLM cost tracking are not commands of Valkey or Redis. On HotKV they are command families of the same server, on the Performance plan and above.

04

Operations tooling outside the engine

Metrics flow to Cloud Monitoring. HotKV serves Prometheus metrics from the server itself when metrics-port is set, and per-command latency histograms through LATENCY HISTOGRAM, wherever it runs.

Measured, not claimed

This Memorystore alternative, measured. HotKV vs Valkey 9.1.2 and Redis 8.10.2 on one machine

HotKV Ltd measured Valkey 9.1.2 and Redis 8.10.2, the open-source engine families behind Memorystore. Both ran as self-hosted single instances on the same machine as HotKV. Google Cloud Memorystore itself was not measured. It offers Valkey 9.1 and Redis 7.x builds of its own, so the exact builds differ from the ones measured. The figures show the engines behind the service, the fair baseline for a Memorystore alternative.

12 of 12
commands where HotKV measured the highest throughput against Valkey
8.6x to 56.6x
HotKV's throughput as a multiple of Valkey's, across the 12 commands it led (MGET to SADD)
None
commands where Valkey measured higher

HotKV Ltd built HotKV from source for this CPU. Valkey ran as its stock release build. Both ran as single instances. Thread counts followed each product's documented all-core setting and were not tuned per product. The gap is smaller on machines with few cores.

SET17.5x
HotKV
50.5M*
Valkey
2.9M
GET22.5x
HotKV
36.6M*
Valkey
1.6M
MGET8.6x
HotKV
4.6M*
Valkey
537K
SETNX29.7x
HotKV
55.3M*
Valkey
1.9M
APPEND19.4x
HotKV
25.1M
Valkey
1.3M
INCR47.3x
HotKV
90.8M
Valkey
1.9M
LPUSH37.5x
HotKV
31.6M
Valkey
842K
RPUSH49.8x
HotKV
43.8M
Valkey
878K
SADD56.6x
HotKV
50.6M*
Valkey
894K
HSET37.1x
HotKV
29.0M*
Valkey
782K
MSET24.3x
HotKV
11.3M
Valkey
465K
PING36.0x
HotKV
125.2M†
Valkey
3.5M

Operations per second, median of 3 runs of 10 seconds. Each pair of bars scales to the faster server on that command.

* Network-limited: the instance's network set this result, so the server's own limit is higher.

† Load-generator-limited: the load-generator machines ran above 85% CPU, so this is their ceiling, not the server's.

Azure D160alds_v7 (AMD EPYC 9V45 (Turin), 160 vCPUs), measured October 9, 2026. 256-byte values, pipeline depth 30, 5,000,000 prefilled keys read uniformly at random. Three dedicated load-generator machines. Every server ran alone as a single instance with persistence off. Each got all cores (Redis 128 I/O threads, its maximum). HotKV Ltd built HotKV from source for this CPU and ran Valkey as its stock release build. Full method and results · Harness and raw data

Side by side

This Memorystore alternative and Google Cloud Memorystore, feature by feature

FeatureHotKVGoogle Cloud Memorystore
EngineHotKV, multi-threaded, C++20Valkey, Redis Cluster, Redis or Memcached
Command executionMulti-threaded. Every core reads, parses and executes commands on one shared storeSingle main thread per node (Valkey, Redis)
ProtocolRedis-compatible: RESP2 and RESP3 with HELLO negotiation. Reports Redis version 8.4.0 to clientsRESP2 and RESP3 (Valkey, Redis); Memcached protocol
Instance modelEvery HotKV Cloud instance is its own isolated machine with dedicated vCores, RAM and diskManaged instances sized in nodes and shards
BillingPer instance, billed monthly or yearly in advance and priced per region. Outbound traffic is bundled, with top-up packs. When a bundle is used up, outbound traffic pauses. HotKV Cloud does not bill overage on outbound trafficPer node (Valkey, Redis Cluster) or per GiB-hour (Redis), in 1-second increments, per the Google Cloud price list
Availability SLANone on Starter, 99.9% on Performance, 99.95% on ScaleUp to 99.99% (Valkey and Redis Cluster, multi-zone with HA)
LLM prompt and semantic cacheBuilt in (PCACHE.*, SCACHE.*), Performance plan and above on HotKV CloudNot an engine feature
AI agent memoryBuilt in (AMEM.*), Performance plan and above on HotKV CloudNot an engine feature
LLM token and cost trackingBuilt in (LLMBUDGET.*, LLMSTATS.*), Performance plan and above on HotKV CloudNot an engine feature
MetricsBuilt in. The server serves /metrics itself once metrics-port is set (on-premise)Cloud Monitoring metrics
Key taggingYes (TAG.*)No
Nanosecond TTLYes (NEXPIRE, NTTL)No. Millisecond TTLs
ClientsRedis and Valkey client libraries, redis-cli, redis-cli --cluster and GUI tools connect without code changes for the core commandsValkey, Redis and Memcached clients

In fairness

Where Google Cloud Memorystore is ahead of this Memorystore alternative, or simply different

This Memorystore alternative page is only useful when it says this part too.

  • Memorystore integrates with Google Cloud IAM, VPC and Cloud Monitoring. Memorystore for Valkey and for Redis Cluster carry a 99.99% availability SLA when deployed multi-zone with high availability. HotKV Cloud's SLA is 99.9% on Performance and 99.95% on Scale, with none on Starter.
  • Memorystore also offers Memcached. HotKV speaks the Redis protocol only.
  • Memorystore for Valkey and for Redis Cluster scale out across zones without downtime. A HotKV Cloud instance is one machine that you upgrade in place.
  • This Memorystore alternative is in early access. Memorystore is a mature service.

Moving over

From Google Cloud Memorystore to this Memorystore alternative in four steps

What switching to a Memorystore alternative involves in practice.

  1. 1

    Create an instance

    Pick a region and a plan in the HotKV Cloud console. The instance is a machine of its own, provisioned, patched and monitored for you.

  2. 2

    Point your clients at it

    Keep the client library you already use. Change the host, port and credentials from Memorystore to the HotKV instance. TLS and password authentication are on from the start, with ACL users available. A Memorystore alternative that speaks the same protocol needs no client changes for the core commands.

  3. 3

    Move the data

    Use the tools you already know. DUMP and RESTORE, MIGRATE, or a replay of your writes all work, because HotKV implements these commands.

  4. 4

    Turn on what was missing

    Vector search and key tagging are core commands on every plan. Prompt and semantic caches, agent memory and LLM cost tracking are commands of the same server on the Performance plan and above. No module, sidecar or second service. That is what a Memorystore alternative should add.

Questions

Questions about this Memorystore alternative

Will my Memorystore clients work with this Memorystore alternative?

Yes, for the Valkey, Redis Cluster and Redis engines. They use RESP2 and RESP3, which HotKV implements. The same client libraries connect to a HotKV instance after a change of host and credentials. Memcached clients use a different protocol and do not.

Did you benchmark Memorystore?

No. HotKV Ltd measured Valkey 9.1.2 and Redis 8.10.2, the open-source engines behind Memorystore. Both ran as self-hosted single instances on the same 160-vCPU machine as HotKV. The managed service and the builds it offers were not measured, so the figures describe the engines.

Is HotKV Cloud in the same regions?

HotKV Cloud is hosted on Google Cloud and offers 42 regions. You pick the region at checkout and the price is shown live for it.

What does a HotKV Cloud instance give me that a Memorystore node does not?

This Memorystore alternative gives you a multi-threaded engine that uses every vCore of its own isolated machine. Vector search, key tagging and nanosecond TTLs on every plan. Prompt and semantic caches, agent memory, RAG and LLM cost tracking from the Performance plan up.

Put this Memorystore alternative to the test with the clients you already have

HotKV Cloud is opening in waves. Every instance of this Memorystore alternative is a dedicated machine with the multi-threaded engine. The AI command families come with the Performance and Scale plans.

Get early access Pricing