Document Cloud RDI workspace capacity planning and pipeline scaling

[ZdravkoDonev-redis]

Cloud RDI pipelines share the workspace's network capacity. This change adds workspace-wide CIDR planning and a processor capacity table for /22, /21, and /20. The table separates configured and Ready replicas. It counts processors only and states the reference configuration and allocation constraints.

The creation guide now treats /22 as the suggested range. It recommends /21 or larger beyond five sources or five processor replicas across all pipelines. These counts are a conservative planning trigger. The guide also explains growth with /20 and keeps the private-range and overlap requirements.

The scaling guide explains bottleneck diagnosis, manual scaling, and running/configured replica mismatches. It links to the customer Cloud API status route. The FAQ explains that a larger CIDR requires workspace recreation and pipeline redeployment. It covers ingestion interruption, new initial snapshots, and support for planning connectivity changes. Instructions use short sentences throughout the changed sections.

Manual verification:

• Verified the Cloud API pipeline-status route in QA. Checked Dashboard labels and configuration properties against the owning UI and service sources.
• Fresh processor-only checks reached 14 Ready replicas from 15 configured with /22, and 32 from 50 with /21. /22 exhausted blocks in the assigned subnet while another zone retained a block. /21 exhausted blocks across all worker subnets and also encountered allocation throttling.
• The earlier /20 check reached all 50 configured replicas. Each workspace had no sources and two control-plane nodes. Temporary workspace scaling overrides enabled the requested counts. These checks establish address-allocation references, not supported replica limits, throughput, combined source/processor capacity, or maintenance headroom.

---

Note

Low Risk
Documentation-only changes to Redis Cloud RDI guides; no application or infrastructure code.

Overview
Adds a new Scale a Cloud RDI pipeline guide and wires it through the RDI docs so customers can plan workspace CIDR up front and scale collectors, the RDI database, and Flink processor replicas deliberately.

The new guide explains shared workspace network capacity (sources + replicas across all pipelines), conservative /22 vs /21 vs /20 planning with an observed processor replica table, bottleneck diagnosis on the Dashboard, tuning steps (including advanced.resources.taskManager.replicas), and how to confirm running vs configured replicas via the console or Cloud API—not billing.

Workspace creation steps are tightened and now call out CIDR sizing before create, the fact that CIDR cannot be enlarged later, and links to the capacity section. The RDI index, FAQ (new capacity/network and processor-scaling Q&A, including workspace recreation for a larger CIDR), and view/edit Settings cross-link to the scaling guide.

^{Reviewed by Cursor Bugbot for commit e83ec5b. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

Staging links:
https://redis.io/docs/staging/cloud-rdi-processor-scaling-docs/operate/rc/rdi/
https://redis.io/docs/staging/cloud-rdi-processor-scaling-docs/operate/rc/rdi/create-workspace
https://redis.io/docs/staging/cloud-rdi-processor-scaling-docs/operate/rc/rdi/faq/
https://redis.io/docs/staging/cloud-rdi-processor-scaling-docs/operate/rc/rdi/quick-start
https://redis.io/docs/staging/cloud-rdi-processor-scaling-docs/operate/rc/rdi/scale-pipeline/
https://redis.io/docs/staging/cloud-rdi-processor-scaling-docs/operate/rc/rdi/view-edit/

[github-actions]

🧠 Redis Memory

Found 18 related items from repository history (1 new this commit):

• PR — RDSC-5718 Update RDI Flink VM and WAIT documentation

  What changed

• PR — RDSC-5718 Update RDI Flink VM and WAIT documentation

  What changed

• PR — DOC-6833: Fix four RDI Cloud doc issues from #rdi-cloud

  • Remove Active-Active from the 'multiple targets' use case sentence; add it as an explicit unsupported target topology under 'When not to u

• PR — RDSC-4603 Update RDI public documentation

  See https://redislabs.atlassian.net/browse/RDSC-4603 and https://github.com/RedisLabs/redis-data-integration/pull/2565

• PR — RDSC-6045: Document draining streams before Flink processor migration

  Switching an existing pipeline from the classic processor to the Flink processor while input streams still contain records can leave records

• Commit — Document Cloud RDI processor scaling

  Document Cloud RDI processor scaling

• Commit — Document Cloud RDI dashboard replica counts

  Document Cloud RDI dashboard replica counts

• Commit — Document Cloud RDI scaling signals

  Document Cloud RDI scaling signals

• Commit — Document Cloud RDI capacity tuning

  Document Cloud RDI capacity tuning

• PR — RDSC-6038: Update Redis Cloud RDI documentation for 2.0.0

  Updates the Cloud RDI guides for the RDI 2.0.0 console. The current guides describe one source and the older five-step setup. The updated gu

• PR — RDSC-5692 Update public docs that Flink processor is GA

  See https://redislabs.atlassian.net/browse/RDSC-5692

• Commit — Clarify Cloud RDI workspace capacity planning and scaling

  Clarify Cloud RDI workspace capacity planning and scaling

• PR — Explain Cloud RDI networking for multiple sources and subscriptions

  Cloud RDI networking docs currently show only one source connection. Expand the existing overview to explain shared and separate PrivateLink

• Commit — Rename Cloud RDI scaling guide

  Rename Cloud RDI scaling guide

• Commit — Clarify default Cloud RDI CIDR capacity

  Clarify default Cloud RDI CIDR capacity

• PR — RDSC-5962: Document AWS CIDR requirements for Cloud RDI workspaces

  Summary

• PR — RC: Note RDI pipeline CIDR requirement for target database allow lists

  Explain that when a target database has a CIDR allow list, the RDI pipeline's CIDR must be added to it or the pipeline can't reach the datab

• Commit — Document observed Cloud RDI processor allocation by CIDR 🆕

  Document observed Cloud RDI processor allocation by CIDR

Memory updated at e83ec5b

[andy-stark-redis]

LGTM

[ZdravkoDonev-redis]

I'm doing additional testing atm to make sure I document everything properly. I'll do last updates today and after that we can merge.