Redis Cloud Pro: Search on Flex Preview, Smooth Scaling and a Datadog Integration

The Five Announcements and Their Availability #

Search on Flex: Search Indexes on SSD #

Search on Flex extends Redis Flex tiering, which already keeps hot keys in RAM and warm ones on SSD, to the search index itself. The full index lives on tiered storage and only hot index metadata stays in RAM, for a memory footprint of roughly ten percent of an in-memory index. Redis states its design target as the "10:10 rule": roughly ten times in-memory query latency at roughly ten percent of the RAM. The RAM-to-Flash ratio can be changed later without re-indexing, and Query Performance Factor (QPF) spreads queries across vCPUs, currently at no additional charge. The workloads Redis expects range from one to the low tens of terabytes.

The preview supports HASH documents, TEXT fields with prefix, infix, suffix, wildcard and fuzzy matching, TAG fields, VECTOR fields with HNSW and FLAT indexes, SORTBY and RETURN, and high availability, persistence, backup and upgrades. NUMERIC and GEO fields, JSON documents, FT.AGGREGATE, FT.HYBRID and background indexing are not yet available on Redis Cloud Pro. Terraform support is at preview level.

Smooth Scaling: Atomic Slot Migration in Redis Cloud Pro #

Smooth Scaling uses atomic slot migration, introduced in Redis 8.4 Open Source, to move only the hash slots that must move, each in a single step. Redis says scaling windows shrink, in some cases by a large factor, with the biggest gain when the requested change is small, and clients see fewer redirects and interruptions; the exact result depends on dataset size, the change and the workload. Scaling stays user-initiated: Smooth Scaling is not automatic or scheduled scaling. Active-Active and Redis Flex databases are not supported yet, and databases using the standard or custom hashing policies keep scaling as before, with no action required.

Datadog Integration in Private Preview #

The Datadog integration streams Redis Cloud metrics to Datadog without a Datadog Agent, VPC peering or firewall changes. It is configured on the Integrations page of the Redis Cloud console with a Datadog API key and region, where US-1 matches app.datadoghq.com and EU-1 matches app.datadoghq.eu. Metrics carry the rdse2. prefix and usually appear within about 30 seconds. Because it is a private preview, access goes through the Redis account team. The Integrations page also lists Confluent and Grafana and Prometheus.

Getting Access to the Previews #

1. For Search on Flex, enable the opt-in preview flag in the Redis Cloud console for a Redis Cloud Pro subscription, then create a Flex database with Search indexes.
2. For Smooth Scaling, no action is needed: once it reaches an account, it applies to eligible Redis 8.4 databases; check that the database uses the default Redis hashing policy if the faster scaling matters.
3. For the Datadog integration, contact the Redis account team, then create a Datadog API key under Organization Settings, API Keys, and connect it from the Integrations page.