Add testing for iterative scheduler backlogged requests

triton-inference-server · Mar 29, 2024 · a9f9c08 · a9f9c08
1 parent 8c3156e
commit a9f9c08
Show file tree

Hide file tree

Showing 2 changed files with 25 additions and 11 deletions.
diff --git a/qa/L0_iterative_sequence/iterative_sequence_e2e.py b/qa/L0_iterative_sequence/iterative_sequence_e2e.py
@@ -46,7 +46,7 @@
 MODEL_CONFIG_BASE = """
 {{
 "backend": "iterative_sequence",
-"max_batch_size": 4,
+"max_batch_size": 1,
 "input" : [
   {{
     "name": "INPUT",
@@ -103,30 +103,44 @@ def test_generate_stream(self):
             self.assertEqual(res_count, data["OUTPUT"])
         self.assertEqual(0, res_count)
 
-    def test_grpc_stream(self, sequence_id=0, sequence_start=False):
+    def test_grpc_stream(
+        self, sequence_id=0, sequence_start=False, num_requests=1, validation=True
+    ):
         user_data = UserData()
         with grpcclient.InferenceServerClient("localhost:8001") as triton_client:
             triton_client.start_stream(callback=partial(callback, user_data))
             inputs = []
             inputs.append(grpcclient.InferInput("INPUT", [1, 1], "INT32"))
             inputs[0].set_data_from_numpy(np.array([[2]], dtype=np.int32))
 
-            triton_client.async_stream_infer(
-                model_name="iterative_sequence",
-                inputs=inputs,
-                sequence_id=sequence_id,
-                sequence_start=sequence_start,
-            )
-            res_count = 2
+            for _ in range(num_requests):
+                triton_client.async_stream_infer(
+                    model_name="iterative_sequence",
+                    inputs=inputs,
+                    sequence_id=sequence_id,
+                    sequence_start=sequence_start,
+                )
+            res_count = 2 * num_requests
             while res_count > 0:
                 data_item = user_data._completed_requests.get()
                 res_count -= 1
                 if type(data_item) == InferenceServerException:
                     raise data_item
                 else:
-                    self.assertEqual(res_count, data_item.as_numpy("OUTPUT")[0][0])
+                    if validation:
+                        self.assertEqual(
+                            res_count % 2, data_item.as_numpy("OUTPUT")[0][0]
+                        )
             self.assertEqual(0, res_count)
 
+    def test_backlog_fill(self):
+        config = r'"sequence_batching" : { "iterative_sequence" : true, "max_sequence_idle_microseconds": 8000000, direct: { "max_queue_delay_microseconds" : 10000000 }}'
+        with grpcclient.InferenceServerClient("localhost:8001") as triton_client:
+            triton_client.load_model(
+                "iterative_sequence", config=MODEL_CONFIG_BASE.format(config)
+            )
+        self.test_grpc_stream(num_requests=4, validation=False)
+
     def test_reschedule_error(self):
         # Use short idle timeout (< backend reschedule delay: 0.5s) so that
         # the backend won't be able to reschedule the request as the scheduler

diff --git a/qa/L0_iterative_sequence/models/iterative_sequence/config.pbtxt b/qa/L0_iterative_sequence/models/iterative_sequence/config.pbtxt
@@ -24,7 +24,7 @@
 # (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 backend: "iterative_sequence"
-max_batch_size: 4
+max_batch_size: 1
 input [
   {
     name: "INPUT"