Skip to content

Commit ab864e0

Browse files
Merge pull request #127 from Rafi-Microsoft/psl-addingfiles
feat: Add quota check scripts for Azure OpenAI models and Fabric capacity
2 parents 2047307 + 605b81e commit ab864e0

File tree

3 files changed

+729
-31
lines changed

3 files changed

+729
-31
lines changed

docs/quota_check.md

Lines changed: 75 additions & 31 deletions
Original file line numberDiff line numberDiff line change
@@ -1,64 +1,97 @@
11
# Check Quota Availability Before Deployment
22

3-
Before deploying the accelerator, **ensure sufficient quota availability** for the required model.
4-
> **We recommend increasing the capacity to 100k tokens for optimal performance.**
3+
Before deploying the accelerator, **ensure sufficient quota availability** for the required AI models and Fabric capacity.
4+
> **The default capacities match the deployment parameters in `infra/main.bicepparam`.**
55
66
## Login if you have not done so already
77
```
88
az login
99
```
1010

1111
## 📌 Default Models & Capacities:
12+
These match the `modelDeploymentList` in the Bicep parameters:
1213
```
13-
gpt-4o:150, gpt-4o-mini:150, gpt-4:150, text-embedding-3-small:100
14+
gpt-4.1-mini:40:GlobalStandard, text-embedding-3-large:40:Standard
1415
```
16+
1517
## 📌 Default Regions:
1618
```
17-
eastus, uksouth, eastus2, northcentralus, swedencentral, westus, westus2, southcentralus, canadacentral, australiaeast, japaneast, norwayeast
19+
eastus, eastus2, swedencentral, uksouth, westus, westus2, southcentralus, canadacentral, australiaeast, japaneast, norwayeast
1820
```
21+
22+
## 📌 Optional: Fabric Capacity Check
23+
The accelerator also deploys a **Microsoft Fabric F8** capacity. Pass `--check-fabric` (bash) or `-CheckFabric` (PowerShell) to verify Fabric SKU availability.
24+
1925
## Usage Scenarios:
2026
- No parameters passed → Default models and capacities will be checked in default regions.
2127
- Only model(s) provided → The script will check for those models in the default regions.
2228
- Only region(s) provided → The script will check default models in the specified regions.
2329
- Both models and regions provided → The script will check those models in the specified regions.
2430
- `--verbose` passed → Enables detailed logging output for debugging and traceability.
31+
- `--check-fabric` passed → Also checks Microsoft Fabric capacity availability.
2532

26-
## **Input Formats**
27-
> Use the --models, --regions, and --verbose options for parameter handling:
33+
## **Input Formats — Bash**
34+
> Use the --models, --regions, --verbose, and --check-fabric options for parameter handling:
2835
29-
✔️ Run without parameters to check default models & regions without verbose logging:
30-
```
31-
./quota_check.sh
36+
✔️ Run without parameters to check default models & regions:
37+
```sh
38+
./quota_check.sh
3239
```
3340
✔️ Enable verbose logging:
34-
```
35-
./quota_check.sh --verbose
41+
```sh
42+
./quota_check.sh --verbose
3643
```
3744
✔️ Check specific model(s) in default regions:
38-
```
39-
./quota_check.sh --models gpt-4o:150,text-embedding-3-small:100
45+
```sh
46+
./quota_check.sh --models gpt-4.1-mini:40:GlobalStandard,text-embedding-3-large:40:Standard
4047
```
4148
✔️ Check default models in specific region(s):
42-
```
43-
./quota_check.sh --regions eastus,westus
44-
```
45-
✔️ Passing Both models and regions:
46-
```
47-
./quota_check.sh --models gpt-4o:150 --regions eastus,westus2
49+
```sh
50+
./quota_check.sh --regions eastus,westus
4851
```
4952
✔️ All parameters combined:
53+
```sh
54+
./quota_check.sh --models gpt-4.1-mini:40 --regions eastus,westus --verbose
55+
```
56+
✔️ Also check Fabric capacity availability:
57+
```sh
58+
./quota_check.sh --check-fabric --verbose
5059
```
51-
./quota_check.sh --models gpt-4:150,text-embedding-3-small:100 --regions eastus,westus --verbose
60+
61+
## **Input Formats — PowerShell**
62+
> Use the -Models, -Regions, -Verbose, and -CheckFabric parameters:
63+
64+
✔️ Run without parameters:
65+
```powershell
66+
.\quota_check.ps1
67+
```
68+
✔️ Check specific model(s):
69+
```powershell
70+
.\quota_check.ps1 -Models "gpt-4.1-mini:40:GlobalStandard,text-embedding-3-large:40:Standard"
5271
```
53-
✔️ Multiple models with single region:
72+
✔️ Check specific region(s):
73+
```powershell
74+
.\quota_check.ps1 -Regions "eastus,westus2"
5475
```
55-
./quota_check.sh --models gpt-4:150,text-embedding-3-small:100 --regions eastus2 --verbose
76+
✔️ All parameters combined:
77+
```powershell
78+
.\quota_check.ps1 -Models "gpt-4.1-mini:40" -Regions "eastus,westus" -CheckFabric -Verbose
5679
```
5780

5881
## **Sample Output**
5982
The final table lists regions with available quota. You can select any of these regions for deployment.
6083

61-
![quota-check-output](../img/Documentation/quota-check-output.png)
84+
```
85+
╔══════════════════════════════════════════════════════════════╗
86+
║ QUOTA CHECK SUMMARY ║
87+
╚══════════════════════════════════════════════════════════════╝
88+
89+
Region gpt-4.1-mini text-embedding-3-large Status
90+
──────────────────────────────────────────────────────────────────────────────────────────
91+
eastus ✅ 200/240 (need 40) ✅ 120/200 (need 40) ✅ PASS
92+
eastus2 ❌ 10/240 (need 40) ✅ 50/200 (need 40) ❌ FAIL
93+
swedencentral ✅ 100/240 (need 40) ✅ 80/200 (need 40) ✅ PASS
94+
```
6295

6396
---
6497
## **If using Azure Portal and Cloud Shell**
@@ -74,22 +107,33 @@ The final table lists regions with available quota. You can select any of these
74107
chmod +x quota_check.sh
75108
./quota_check.sh
76109
```
77-
- Refer to [Input Formats](#input-formats) for detailed commands.
110+
- Refer to [Input Formats — Bash](#input-formats--bash) for detailed commands.
78111

79112
## **If using VS Code or Codespaces**
113+
114+
### Option 1: Bash (Linux, macOS, Git Bash, WSL, Cloud Shell)
80115
1. Open the terminal in VS Code or Codespaces.
81-
2. Use a terminal that can run bash. This is only for the quota check script; deployment uses PowerShell.
116+
2. Use a terminal that can run bash.
82117
![git_bash](../img/provisioning/git_bash.png)
83-
3. Navigate to the `scripts` folder where the script files are located and make the script as executable:
118+
3. Navigate to the `scripts` folder and make the script executable:
84119
```sh
85120
cd scripts
86121
chmod +x quota_check.sh
87122
```
88-
4. Run the appropriate script based on your requirement:
89-
90-
**To check quota for the deployment**
91-
123+
4. Run the script:
92124
```sh
93125
./quota_check.sh
94126
```
95-
- Refer to [Input Formats](#input-formats) for detailed commands.
127+
- Refer to [Input Formats — Bash](#input-formats--bash) for detailed commands.
128+
129+
### Option 2: PowerShell (Windows, Linux, macOS)
130+
1. Open a PowerShell terminal in VS Code.
131+
2. Navigate to the `scripts` folder:
132+
```powershell
133+
cd scripts
134+
```
135+
3. Run the script:
136+
```powershell
137+
.\quota_check.ps1
138+
```
139+
- Refer to [Input Formats — PowerShell](#input-formats--powershell) for detailed commands.

0 commit comments

Comments
 (0)