[Refactor]Refactor of vllm_ascend/distributed module (#5719)
### What this PR does / why we need it?
Based on the RFC:https://github.com/vllm-project/vllm-ascend/issues/5604
This PR is a refactoring of vllm_ascend/distributed, moving all
kv_transfer realtaed codes into a dedicated folder, which has already
been done in vLLM
### Does this PR introduce _any_ user-facing change?
NA
### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef
---------
Signed-off-by: lty <linhebiwen@gmail.com>
This commit is contained in:
@@ -43,7 +43,6 @@ deployment:
|
||||
"kv_role": "kv_producer",
|
||||
"kv_port": "30000",
|
||||
"engine_id": "0",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
@@ -81,7 +80,6 @@ deployment:
|
||||
"kv_role": "kv_producer",
|
||||
"kv_port": "30100",
|
||||
"engine_id": "1",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
@@ -120,7 +118,6 @@ deployment:
|
||||
"kv_role": "kv_consumer",
|
||||
"kv_port": "30200",
|
||||
"engine_id": "2",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
@@ -158,7 +155,6 @@ deployment:
|
||||
"kv_role": "kv_consumer",
|
||||
"kv_port": "30200",
|
||||
"engine_id": "2",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
|
||||
@@ -46,7 +46,6 @@ deployment:
|
||||
"kv_role": "kv_producer",
|
||||
"kv_port": "30000",
|
||||
"engine_id": "0",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 1,
|
||||
@@ -85,7 +84,6 @@ deployment:
|
||||
"kv_role": "kv_consumer",
|
||||
"kv_port": "30100",
|
||||
"engine_id": "1",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 1,
|
||||
|
||||
@@ -44,7 +44,6 @@ deployment:
|
||||
"kv_role": "kv_producer",
|
||||
"kv_port": "30000",
|
||||
"engine_id": "0",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
@@ -83,7 +82,6 @@ deployment:
|
||||
"kv_role": "kv_producer",
|
||||
"kv_port": "30100",
|
||||
"engine_id": "1",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
@@ -124,7 +122,6 @@ deployment:
|
||||
"kv_role": "kv_consumer",
|
||||
"kv_port": "30200",
|
||||
"engine_id": "2",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
@@ -168,7 +165,6 @@ deployment:
|
||||
"kv_role": "kv_consumer",
|
||||
"kv_port": "30200",
|
||||
"engine_id": "2",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
|
||||
@@ -53,7 +53,6 @@ deployment:
|
||||
"kv_role": "kv_producer",
|
||||
"kv_port": "30000",
|
||||
"engine_id": "0",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
@@ -88,7 +87,6 @@ deployment:
|
||||
"kv_role": "kv_consumer",
|
||||
"kv_port": "30200",
|
||||
"engine_id": "1",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
|
||||
@@ -39,7 +39,6 @@ deployment:
|
||||
"kv_role": "kv_producer",
|
||||
"kv_port": "30000",
|
||||
"engine_id": "0",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
@@ -76,7 +75,6 @@ deployment:
|
||||
"kv_role": "kv_consumer",
|
||||
"kv_port": "30200",
|
||||
"engine_id": "1",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
|
||||
@@ -43,7 +43,6 @@ deployment:
|
||||
"kv_role": "kv_producer",
|
||||
"kv_port": "30000",
|
||||
"engine_id": "0",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 1,
|
||||
@@ -83,7 +82,6 @@ deployment:
|
||||
"kv_role": "kv_consumer",
|
||||
"kv_port": "30100",
|
||||
"engine_id": "1",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 1,
|
||||
|
||||
@@ -39,7 +39,6 @@ deployment:
|
||||
"kv_role": "kv_producer",
|
||||
"kv_port": "30000",
|
||||
"engine_id": "0",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
@@ -74,7 +73,6 @@ deployment:
|
||||
"kv_role": "kv_consumer",
|
||||
"kv_port": "30200",
|
||||
"engine_id": "1",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
|
||||
@@ -44,7 +44,6 @@ deployment:
|
||||
"kv_role": "kv_producer",
|
||||
"kv_port": "30000",
|
||||
"engine_id": "0",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"use_ascend_direct": true,
|
||||
"prefill": {
|
||||
@@ -84,7 +83,6 @@ deployment:
|
||||
"kv_role": "kv_consumer",
|
||||
"kv_port": "30100",
|
||||
"engine_id": "1",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"use_ascend_direct": true,
|
||||
"prefill": {
|
||||
|
||||
@@ -39,7 +39,6 @@ deployment:
|
||||
"kv_role": "kv_producer",
|
||||
"kv_port": "30000",
|
||||
"engine_id": "0",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
@@ -73,7 +72,6 @@ deployment:
|
||||
"kv_role": "kv_consumer",
|
||||
"kv_port": "30200",
|
||||
"engine_id": "1",
|
||||
"kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
|
||||
"kv_connector_extra_config": {
|
||||
"prefill": {
|
||||
"dp_size": 2,
|
||||
|
||||
Reference in New Issue
Block a user