Conversation
Add r6i and e2-highmem pricing entries, merge matching AWS/GCP instance lines, and include cores and memory columns in standalone and replicated tables for readability. Co-Authored-By: Oz <oz-agent@warp.dev>
📝 WalkthroughWalkthroughUpdated documentation and token list: pricing tables in Changes
Estimated Code Review Effort🎯 2 (Simple) | ⏱️ ~10 minutes Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Add highmem and <htmlcontent> to the shared wordlist to resolve spellcheck failures triggered by the Pro tier pricing table updates. Co-Authored-By: Oz <oz-agent@warp.dev>
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
cloud/pro-tier.md (1)
72-72: Consider documenting resource overhead in operations guide.The note mentions that replication requires "an additional 2 cores and 2 GB" for infrastructure components (sentinel and rebalancer). This operational detail would be valuable to include in the
operations/replication.mdguide, which currently focuses on setup but doesn't mention resource requirements.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@cloud/pro-tier.md` at line 72, Add the replication resource-overhead note from cloud/pro-tier.md into the operations guide by documenting that replication and cluster modes require an extra 2 cores and 2 GB per instance to run infrastructure components (sentinel and rebalancer); update operations/replication.md to include a short "Resource requirements" section that references Cores and Memory (GB) as per-instance values and explicitly states the additional 2 cores/2 GB overhead for sentinel and rebalancer so operators can provision capacity accordingly.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@cloud/pro-tier.md`:
- Around line 47-49: Spellchecker incorrectly flags the GCP instance family
token "highmem"; add the token "highmem" to the project's spellcheck dictionary
(the custom word list used by the CI spellchecker) so the pipeline accepts
entries like "E2-highmem-2". Update the dictionary file to include the lowercase
"highmem" (and any case variants if your checker is case-sensitive), commit the
change, and re-run the pipeline to confirm the spellcheck warning is resolved.
---
Nitpick comments:
In `@cloud/pro-tier.md`:
- Line 72: Add the replication resource-overhead note from cloud/pro-tier.md
into the operations guide by documenting that replication and cluster modes
require an extra 2 cores and 2 GB per instance to run infrastructure components
(sentinel and rebalancer); update operations/replication.md to include a short
"Resource requirements" section that references Cores and Memory (GB) as
per-instance values and explicitly states the additional 2 cores/2 GB overhead
for sentinel and rebalancer so operators can provision capacity accordingly.
| | E2-highmem-2 / r6i.large | 2 | 16 | $408.80 | | ||
| | E2-highmem-4 / r6i.xlarge | 4 | 32 | $817.60 | | ||
| | E2-highmem-8 / r6i.2xlarge | 8 | 64 | $1,635.20 | |
There was a problem hiding this comment.
Address spellcheck failure: add "highmem" to dictionary.
The pipeline is flagging "highmem" as misspelled, but this is the official GCP naming convention for high-memory instance families (e.g., E2-highmem-2). Add "highmem" to the spellcheck dictionary to resolve this failure.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@cloud/pro-tier.md` around lines 47 - 49, Spellchecker incorrectly flags the
GCP instance family token "highmem"; add the token "highmem" to the project's
spellcheck dictionary (the custom word list used by the CI spellchecker) so the
pipeline accepts entries like "E2-highmem-2". Update the dictionary file to
include the lowercase "highmem" (and any case variants if your checker is
case-sensitive), commit the change, and re-run the pipeline to confirm the
spellcheck warning is resolved.
|
@oz-agent the supported instance types:
|
Add e2-medium, t2.medium, e2-standard-8/m6i.2xlarge, m6i.4xlarge, and m6i.8xlarge to standalone and replicated pricing tables with cores, memory, and monthly costs. Co-Authored-By: Oz <oz-agent@warp.dev>
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
cloud/pro-tier.md (1)
82-82: Clarify the replication note to avoid confusion with cluster deployment.The note mentions "replication and cluster" and references a "rebalancer for cluster," but this table section is specifically titled "Replicated (High Availability, Master (x1), Replica (x1))" with no cluster deployment. The mention of cluster components in a replication-only context may confuse readers.
Consider revising the note to clearly separate replication overhead from cluster overhead, or remove the cluster reference if it doesn't apply to this pricing table.
📝 Suggested clarification
-> Note: In the Replicated table, **Cores** and **Memory (GB)** are per instance. We charge an additional 2 cores and 2 GB for replication and cluster since they require an extra component (sentinel for replication and rebalancer for cluster). +> Note: In the Replicated table, **Cores** and **Memory (GB)** are per instance. We charge an additional 2 cores and 2 GB for replication overhead, as it requires an extra component (sentinel) to manage high availability.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@cloud/pro-tier.md` at line 82, The note sentence "Note: In the Replicated table, **Cores** and **Memory (GB)** are per instance. We charge an additional 2 cores and 2 GB for replication and cluster since they require an extra component (sentinel for replication and rebalancer for cluster)." is confusing in the Replicated section; update it to separate replication overhead from cluster overhead (or remove the cluster part if cluster does not apply): explicitly state that the Replicated table charges an additional 2 cores and 2 GB to cover the replication sentinel only, and add a separate sentence (or a cross-reference) that cluster deployments incur their own extra resources (rebalancer) if and only if cluster pricing is presented elsewhere.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@cloud/pro-tier.md`:
- Around line 43-59: Change all occurrences of capitalized GCP E2 instance names
to lowercase to match official docs: replace "E2-medium", "E2-standard-2",
"E2-standard-4", "E2-standard-8", "E2-highmem-2", "E2-highmem-4",
"E2-highmem-8", and the "E2-custom-*-*" entries (e.g., "E2-custom-4-8192",
"E2-custom-8-16384", "E2-custom-16-32768", "E2-custom-32-65536") with
"e2-medium", "e2-standard-2", "e2-standard-4", "e2-standard-8", "e2-highmem-2",
"e2-highmem-4", "e2-highmem-8", and the corresponding "e2-custom-*-*" names
(also update "E2-standard-2 / m6i.large (Starting Instance)" to "e2-standard-2 /
m6i.large (Starting Instance)"); leave other provider names (m6i, r6i, c6i)
unchanged.
---
Nitpick comments:
In `@cloud/pro-tier.md`:
- Line 82: The note sentence "Note: In the Replicated table, **Cores** and
**Memory (GB)** are per instance. We charge an additional 2 cores and 2 GB for
replication and cluster since they require an extra component (sentinel for
replication and rebalancer for cluster)." is confusing in the Replicated
section; update it to separate replication overhead from cluster overhead (or
remove the cluster part if cluster does not apply): explicitly state that the
Replicated table charges an additional 2 cores and 2 GB to cover the replication
sentinel only, and add a separate sentence (or a cross-reference) that cluster
deployments incur their own extra resources (rebalancer) if and only if cluster
pricing is presented elsewhere.
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: d827aaa3-0313-4e11-a2f4-8a0c57c37d5c
📒 Files selected for processing (2)
.wordlist.txtcloud/pro-tier.md
✅ Files skipped from review due to trivial changes (1)
- .wordlist.txt
| | Instance Type | Cores | Memory (GB) | Monthly Cost | | ||
| | :--- | :---: | :---: | ---: | | ||
| | E2-medium | 1 | 4 | $175.20 | | ||
| | t2.medium | 2 | 4 | $321.20 | | ||
| | E2-standard-2 / m6i.large (Starting Instance) | 2 | 8 | $350.40 | | ||
| | E2-standard-4 / m6i.xlarge | 4 | 16 | $700.80 | | ||
| | E2-standard-8 / m6i.2xlarge | 8 | 32 | $1,401.60 | | ||
| | E2-highmem-2 / r6i.large | 2 | 16 | $408.80 | | ||
| | E2-highmem-4 / r6i.xlarge | 4 | 32 | $817.60 | | ||
| | E2-highmem-8 / r6i.2xlarge | 8 | 64 | $1,635.20 | | ||
| | m6i.4xlarge | 16 | 64 | $2,803.20 | | ||
| | m6i.8xlarge | 32 | 128 | $5,606.40 | | ||
| | r6i.4xlarge | 16 | 128 | $3,270.40 | | ||
| | E2-custom-4-8192 / c6i.xlarge | 4 | 8 | $642.40 | | ||
| | E2-custom-8-16384 / c6i.2xlarge | 8 | 16 | $1,284.80 | | ||
| | E2-custom-16-32768 / c6i.4xlarge | 16 | 32 | $2,569.60 | | ||
| | E2-custom-32-65536 / c6i.8xlarge | 32 | 64 | $5,139.20 | |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
# Check GCP instance naming patterns in the file
echo "=== GCP Instance Naming in File ==="
grep -n "E2-\|e2-" cloud/pro-tier.md | head -20
echo -e "\n=== File structure around lines 43-59 ==="
sed -n '40,65p' cloud/pro-tier.mdRepository: FalkorDB/docs
Length of output: 2636
🌐 Web query:
What are the official specifications for GCP e2 instance types (e2-medium, e2-standard-2, e2-standard-4, e2-standard-8, e2-highmem-2, e2-highmem-4, e2-highmem-8)?
💡 Result:
Official Compute Engine E2 machine-type specifications (predefined), from Google Cloud docs: vCPUs, memory, max # of Persistent Disks (PDs), max total PD size, Local SSD support, and max egress bandwidth. [1]
| Machine type | vCPUs | Memory (GB) | Max # PDs | Max total PD size (TiB) | Local SSD | Max egress bandwidth (Gbps) |
|---|---|---|---|---|---|---|
e2-medium |
2 (shared-core; 1 fractional vCPU) | 4 | 16 | 3 | No | Up to 2 |
e2-standard-2 |
2 | 8 | 128 | 257 | No | Up to 4 |
e2-standard-4 |
4 | 16 | 128 | 257 | No | Up to 8 |
e2-standard-8 |
8 | 32 | 128 | 257 | No | Up to 16 |
e2-highmem-2 |
2 | 16 | 128 | 257 | No | Up to 4 |
e2-highmem-4 |
4 | 32 | 128 | 257 | No | Up to 8 |
e2-highmem-8 |
8 | 64 | 128 | 257 | No | Up to 16 |
Source: Google Cloud “General-purpose machine family for Compute Engine” (E2 machine series tables). [1]
🌐 Web query:
What are the official specifications (vCPUs and memory) for AWS instance types mentioned in pricing tables (t2.medium, m6i.large, m6i.xlarge, m6i.2xlarge, m6i.4xlarge, m6i.8xlarge, r6i.large, r6i.xlarge, r6i.2xlarge, r6i.4xlarge, c6i.xlarge, c6i.2xlarge, c6i.4xlarge, c6i.8xlarge)?
💡 Result:
| Instance type | vCPUs | Memory (GiB) | Source |
|---|---|---|---|
| t2.medium | 2 | 4 | [1] |
| m6i.large | 2 | 8 | [1] |
| m6i.xlarge | 4 | 16 | [1] |
| m6i.2xlarge | 8 | 32 | [1] |
| m6i.4xlarge | 16 | 64 | [1] |
| m6i.8xlarge | 32 | 128 | [1] |
| r6i.large | 2 | 16 | [2] |
| r6i.xlarge | 4 | 32 | [2] |
| r6i.2xlarge | 8 | 64 | [2] |
| r6i.4xlarge | 16 | 128 | [2] |
| c6i.xlarge | 4 | 8 | [3] |
| c6i.2xlarge | 8 | 16 | [3] |
| c6i.4xlarge | 16 | 32 | [3] |
| c6i.8xlarge | 32 | 64 | [3] |
Sources
[1] AWS EC2 General purpose instance types (tables for M6i and T2)
[2] AWS EC2 Memory optimized instance types (table for R6i)
[3] AWS EC2 Compute optimized instance types (table for C6i)
🌐 Web query:
Is "highmem" the correct naming for GCP memory-optimized instance types or should it be "highmem"?
💡 Result:
In Google Compute Engine machine type names, the ratio keyword is highmem (no dash), e.g. n2-highmem-8. [1]
However, memory-optimized Compute Engine machine types are not named *highmem*; they’re in the memory-optimized families/series (for example M1/M2/M3/M4 and X4) and use names like m3-ultramem-*, m3-megamem-*, m1-ultramem-*, etc. [2]
Sources: [1], [2]
Use lowercase GCP instance type names to match official documentation.
The table uses capitalized "E2-" for GCP instances (e.g., "E2-medium", "E2-standard-2"), but GCP's official documentation consistently uses lowercase "e2-" (e.g., "e2-medium", "e2-standard-2"). Change all capitalized E2 instance types to lowercase e2 for consistency with GCP's official naming convention.
All instance specifications (cores and memory) have been verified against official cloud provider documentation and are correct.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@cloud/pro-tier.md` around lines 43 - 59, Change all occurrences of
capitalized GCP E2 instance names to lowercase to match official docs: replace
"E2-medium", "E2-standard-2", "E2-standard-4", "E2-standard-8", "E2-highmem-2",
"E2-highmem-4", "E2-highmem-8", and the "E2-custom-*-*" entries (e.g.,
"E2-custom-4-8192", "E2-custom-8-16384", "E2-custom-16-32768",
"E2-custom-32-65536") with "e2-medium", "e2-standard-2", "e2-standard-4",
"e2-standard-8", "e2-highmem-2", "e2-highmem-4", "e2-highmem-8", and the
corresponding "e2-custom-*-*" names (also update "E2-standard-2 / m6i.large
(Starting Instance)" to "e2-standard-2 / m6i.large (Starting Instance)"); leave
other provider names (m6i, r6i, c6i) unchanged.
2 participants
Summary
Co-Authored-By: Oz oz-agent@warp.dev
Summary by CodeRabbit