Support data parallelism (static) (#480)

Co-authored-by: Ying Sheng <ying.sheng@databricks.com>
Co-authored-by: Lianmin Zheng <lianminzheng@gmail.com>
Co-authored-by: Liangsheng Yin <hnyls2002@gmail.com>
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>

python/sglang/global_config.py

@@ -26,7 +26,8 @@ class GlobalConfig:
         self.concate_and_append_mode = "no_adjust"
 
         # Request dependency time due to network delay
-        self.request_dependency_time = 0.03
+        self.request_dependency_delay = 0.03
+        self.wait_for_new_request_delay = 0.0006
 
         # New generation token ratio estimation
         self.base_new_token_ratio = 0.4