Use monotonic time where appropriate (#1249)

vllm/engine/llm_engine.py

@@ -256,10 +256,10 @@ class LLMEngine:
             prompt_token_ids: The token IDs of the prompt. If None, we
                 use the tokenizer to convert the prompts to token IDs.
             arrival_time: The arrival time of the request. If None, we use
-                the current time.
+                the current monotonic time.
         """
         if arrival_time is None:
-            arrival_time = time.time()
+            arrival_time = time.monotonic()
         if prompt_token_ids is None:
             assert prompt is not None
             prompt_token_ids = self.tokenizer.encode(prompt)
@@ -568,7 +568,7 @@ class LLMEngine:
         prompt_run: bool,
         num_batched_tokens: int,
     ) -> None:
-        now = time.time()
+        now = time.monotonic()
         # Log the number of batched input tokens.
         if prompt_run:
             self.num_prompt_tokens.append((now, num_batched_tokens))