Question 1

What does Redis consulting include?

Accepted Answer

Our Redis consulting covers product-strategy decisions (Redis OSS vs Redis Enterprise vs Redis Stack vs ElastiCache vs MemoryDB), module-architecture review (RediSearch, RedisJSON, RedisTimeSeries, vector / RedisVL), Sentinel vs Cluster topology selection, persistence trade-offs (RDB / AOF / hybrid), licensing-impact analysis under SSPL/RSAL, and a written remediation roadmap. We deliver decisions and trade-off documentation — execution (migrations, tuning rollout) is a separate engagement.

Question 2

Should I stay on Redis or migrate to Valkey?

Accepted Answer

If your distribution model is unaffected by SSPL/RSAL — most internal-only deployments are — staying on Redis 7.4+ is the path of least disruption, especially if you depend on commercial modules (RediSearch, RedisJSON, RedisGraph) or Redis Enterprise features (Active-Active CRDTs, Auto Tiering, Search 2.x). If you're a hosting provider, a SaaS that bundles Redis with your distribution, or you're on ElastiCache and want the 20% Valkey discount, the migration math changes. We build the decision for your specific deployment model — see also our reverse perspective on the dedicated Valkey consulting page.

Question 3

Redis Enterprise vs Redis OSS — which should I run?

Accepted Answer

Redis OSS covers ~80% of workloads at zero license cost. Redis Enterprise pays for itself when you specifically need: Active-Active CRDTs across regions, Auto Tiering (cold data on NVMe, hot in RAM), Redis on Flash, multi-tenancy with strong isolation, or 24×7 vendor support with single-throat-to-choke accountability. We model the actual delta against your workload — not Redis Inc.'s marketing comparison — and surface whether the Enterprise premium is justified.

Question 4

ElastiCache, MemoryDB, or self-managed on EC2 — what's the right answer?

Accepted Answer

ElastiCache Redis is the default for ephemeral caching where data loss is recoverable. MemoryDB is the right answer when you need durable Redis-compatible storage as a primary database (multi-AZ transactional log, recovery to a known state). Self-managed on EC2 wins on cost at scale (50+ node-hours/day) but absorbs operational burden including upgrades, patching, monitoring, and on-call. We build the actual ROI model for your usage — not just sticker-price arithmetic.

Question 5

When should I add RediSearch, RedisJSON, or vector search via RedisVL?

Accepted Answer

Add RediSearch when full-text search latency requirements (p99 < 5ms) make Elasticsearch overkill or when your search index fits in memory. Add RedisJSON when your access pattern is read-heavy lookups on document fragments — it beats both fetching whole JSON blobs and using HSET for nested structures. Add vector search (RedisVL / Search 2.x) for retrieval-augmented generation (RAG) and semantic-similarity workloads where latency matters more than recall ceiling vs. a dedicated vector DB. We audit your existing query patterns before recommending any module — modules add operational surface.

Question 6

How do you size Redis memory and pick an eviction policy?

Accepted Answer

Sizing is working-set + key-space overhead (~20-30% of dataset) + replication buffer (1-2× max single-transaction size) + client output buffer headroom. Eviction policy depends on access patterns: allkeys-lru for catalog/session caches with skewed access, volatile-ttl for time-bounded data, allkeys-lfu when recency matters less than frequency, noeviction when Redis is the source of truth and OOM should fail the write. We model these against your traffic and surface the right combination rather than recommending defaults.

Question 7

Does Redis consulting include migration execution?

Accepted Answer

No — execution lives on dedicated pages and is a separate engagement shape. Consulting is the decision phase: choose the topology, pick the product tier, model the cost, write the runbook outline. Execution covers migrating ElastiCache → MemoryDB, rolling Redis 6 → 7.4, sharding a single instance to Cluster mode, or moving to Valkey. The split keeps consulting deliverables crisp — written recommendations and trade-off analyses — without scope creep into code-shipping work.

Product	Fits when	Avoid when
Redis OSS (7.4+)	Standard caching, Sentinel HA, cluster sharding, self-managed control.	SSPL/RSAL conflicts with your distribution model, or you need Active-Active.
Redis Stack	You need RediSearch, RedisJSON, RedisTimeSeries, or vector search bundled.	Pure-caching workload — Stack adds modules you'll never use.
Redis Enterprise	Active-Active CRDTs, Auto Tiering, Redis on Flash, vendor SLA, multi-tenant isolation.	Smaller deployments where the license premium isn't offset by feature use.
ElastiCache	Ephemeral cache on AWS, ops outsourcing acceptable, latency-tolerant of cross-AZ.	Durable storage requirements — that's MemoryDB territory.
MemoryDB	Durable Redis-compatible storage as primary DB, multi-AZ tx log, recovery to known state.	Ephemeral cache where ElastiCache's per-hour cost is hard to beat.
Valkey	SSPL/RSAL conflicts, distribution-model concerns, ElastiCache cost reduction (~20%).	You rely on commercial Redis modules or Redis Enterprise features.

Redis Consulting Services

What our Redis consulting covers

How a Redis consulting engagement is shaped

Redis product matrix — when each one fits

Redis consulting — common questions

Ready to make the call on Redis?