microsoft / microsoft/simplechat
Update Readme with scaling and limits details
Open
@paullizer is already working on this.
Since Apr 30, 2025.
documentation
- Dominant language
- Python
- Stars
- 152
- Forks
- 116
- Avg merge
- 7h 7m
- Merged PRs (30d)
- 122
Description
AI Search (S1 Tier)
- Supports up to 24 billion documents across Basic, S1, S2, and S3 tiers.
- On a newly created S1 service (in regions that support higher-capacity partitions), you can scale to 12 partitions at 160 GB each for a total of ~1.92 TB.
- While individual document chunks in the index are limited to ~16 MB, the effective average max document chunk size is in the index is ~88 KB based on max scale.
- In real-world usage (e.g., Simple Chat RAG), document chunks are typically around 34 KB.
- Assuming 20-page documents (~88 KB for each page aka chunk), you can index up to ~1.2 million Office files (PDF, Word, PowerPoint, etc.) per index (user or group).
⸻
Cosmos DB (Provisioned Throughput Mode)
• Container-Level Storage:
• Each container can scale to unlimited storage, distributed across logical partitions (each capped at 20 GB).
• Database-Level Storage:
• Databases with shared throughput have no fixed storage limit; you can add more containers and data as needed.
• Account-Level Limits:
• Up to 500 databases/containers per account.
• Each container supports unlimited storage, adhering to the 20 GB per logical partition constraint.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.