Skip to content

Commit e990869

Browse files
committed
add docs and improve naming
Signed-off-by: Ning Wang <[email protected]>
1 parent aa3ee3d commit e990869

File tree

3 files changed

+21
-6
lines changed

3 files changed

+21
-6
lines changed
Lines changed: 15 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,15 @@
1+
# Dynamo Installation Instruction
2+
3+
4+
We follow the instruction in [dynamo](https://github.com/ai-dynamo/dynamo) to deploy the Dynamo Cloud in Kubernetes. The detailed instrunction can be found in Section 1. `1. Installing Dynamo Cloud from Published Artifacts` from dynamo's [quickstart guide](https://github.com/ai-dynamo/dynamo/blob/main/docs/guides/dynamo_deploy/quickstart.md). We use the most recent release images (version: 0.3.2) published by Dynamo team.
5+
6+
7+
### Model Deployment
8+
9+
We use sample deployment yamls from the dynamo repo in the v0.3.2 release for PD disaggration testing. https://github.com/ai-dynamo/dynamo/blob/v0.3.2/examples/llm/deploy/agg.yaml and https://github.com/ai-dynamo/dynamo/blob/v0.3.2/examples/llm/deploy/agg-router.yaml.
10+
11+
12+
> Note: There are some configuration changes in terms of image downloading and model downloading due to the testing environment difference.
13+
14+
> 1. We download container image from VKE docker registry aibrix-cn-beijing.cr.volces.com. The images are synced from dockerhub and nvidia ngc.
15+
> 2. We download model from VKE object storage, which are synced from Huggingface model hub.

test/regression/v0.4.0/sglang/qwen-8b/aibrix-router-1p1d.yaml

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
apiVersion: orchestration.aibrix.ai/v1alpha1
22
kind: StormService
33
metadata:
4-
name: aibrix-router-1p1d
4+
name: aibrix-sglang-1p1d
55
namespace: default
66
spec:
77
replicas: 1
@@ -10,11 +10,11 @@ spec:
1010
stateful: true
1111
selector:
1212
matchLabels:
13-
app: aibrix-router-1p1d
13+
app: aibrix-sglang-1p1d
1414
template:
1515
metadata:
1616
labels:
17-
app: aibrix-router-1p1d
17+
app: aibrix-sglang-1p1d
1818
spec:
1919
roles:
2020
- name: prefill

test/regression/v0.4.0/sglang/qwen-8b/aibrix-router-4p3d.yaml

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
apiVersion: orchestration.aibrix.ai/v1alpha1
22
kind: StormService
33
metadata:
4-
name: aibrix-router-4p3d
4+
name: aibrix-sglang-4p3d
55
namespace: default
66
spec:
77
replicas: 1
@@ -10,11 +10,11 @@ spec:
1010
stateful: true
1111
selector:
1212
matchLabels:
13-
app: aibrix-router-4p3d
13+
app: aibrix-sglang-4p3d
1414
template:
1515
metadata:
1616
labels:
17-
app: aibrix-router-4p3d
17+
app: aibrix-sglang-4p3d
1818
spec:
1919
roles:
2020
- name: prefill

0 commit comments

Comments
 (0)