From cac6c3333127fd5007d93663664b5b5da1538e86 Mon Sep 17 00:00:00 2001
From: Craig Taverner <craig@amanzi.com>
Date: Tue, 19 Sep 2023 21:08:57 +0200
Subject: [PATCH] Refined ESQL benchmarks, fixed NaN and removed segment (#457)

* Fixed NaN results in `avg_tip_percent_esql` query
* Removed all `segment` partitioning operations since they were unused
* Added `setup` tags to the re-indexing to allow benchmark-only runs with `--exclude-tasks="tag:setup`
---
 nyc_taxis/challenges/default.json | 15 ++++--
 nyc_taxis/operations/default.json | 84 ++++---------------------------
 2 files changed, 19 insertions(+), 80 deletions(-)

diff --git a/nyc_taxis/challenges/default.json b/nyc_taxis/challenges/default.json
index e16ee3f3..3f9c6ee7 100644
--- a/nyc_taxis/challenges/default.json
+++ b/nyc_taxis/challenges/default.json
@@ -710,7 +710,8 @@
       "default": false,
       "schedule": [
         {
-          "operation": "delete-index"
+          "operation": "delete-index",
+          "tags": ["setup"]
         },
         {
           "operation": {
@@ -720,7 +721,8 @@
               "index.refresh_interval": "30s",
               "index.translog.flush_threshold_size": "4g"
             }{%- endif %}
-          }
+          },
+          "tags": ["setup"]
         },
         {
           "name": "check-cluster-health",
@@ -732,17 +734,20 @@
               "wait_for_no_relocating_shards": "true"
             },
             "retry-until-success": true
-          }
+          },
+          "tags": ["setup"]
         },
         {
           "operation": "index",
           "warmup-time-period": 240,
           "clients": {{bulk_indexing_clients | default(8)}},
-          "ignore-response-error-level": "{{error_level | default('non-fatal')}}"
+          "ignore-response-error-level": "{{error_level | default('non-fatal')}}",
+          "tags": ["setup"]
         },
         {
           "name": "refresh-after-index",
-          "operation": "refresh"
+          "operation": "refresh",
+          "tags": ["setup"]
         },
         {
           "operation": "avg_passenger_count_aggregation",
diff --git a/nyc_taxis/operations/default.json b/nyc_taxis/operations/default.json
index 89c13231..aa3c16f0 100644
--- a/nyc_taxis/operations/default.json
+++ b/nyc_taxis/operations/default.json
@@ -695,18 +695,6 @@
         }
       }
     },
-    {
-      "name": "avg_passenger_count_esql_segment_partitioning",
-      "operation-type": "raw-request",
-      "method": "POST",
-      "path": "/_esql",
-      "body": {
-        "query" : "from nyc_taxis | stats avg(passenger_count)",
-        "pragma" : {
-          "data_partitioning" : "segment"
-        }
-      }
-    },
     {
       "name": "avg_passenger_count_esql_doc_partitioning",
       "operation-type": "raw-request",
@@ -725,6 +713,13 @@
       "request-timeout": 120,
       "body": {
         "size": 0,
+        "query": {
+          "range": {
+            "fare_amount": {
+            "gt": 0
+            }
+          }
+        },
         "runtime_mappings": {
           "tip_percent": {
             "type": "double",
@@ -748,31 +743,19 @@
       "method": "POST",
       "path": "/_esql",
       "body": {
-        "query" : "from nyc_taxis | eval tip_percent = tip_amount / fare_amount | stats avg(tip_percent)",
+        "query" : "from nyc_taxis | where fare_amount > 0 | eval tip_percent = tip_amount / fare_amount | stats avg(tip_percent)",
         "pragma" : {
           "data_partitioning" : "shard"
         }
       }
     },
-    {
-      "name": "avg_tip_percent_esql_segment_partitioning",
-      "operation-type": "raw-request",
-      "method": "POST",
-      "path": "/_esql",
-      "body": {
-        "query" : "from nyc_taxis | eval tip_percent = tip_amount / fare_amount | stats avg(tip_percent)",
-        "pragma" : {
-          "data_partitioning" : "segment"
-        }
-      }
-    },
     {
       "name": "avg_tip_percent_esql_doc_partitioning",
       "operation-type": "raw-request",
       "method": "POST",
       "path": "/_esql",
       "body": {
-        "query" : "from nyc_taxis | eval tip_percent = tip_amount / fare_amount | stats avg(tip_percent)",
+        "query" : "from nyc_taxis | where fare_amount > 0 | eval tip_percent = tip_amount / fare_amount | stats avg(tip_percent)",
         "pragma" : {
           "data_partitioning" : "doc"
         }
@@ -815,18 +798,6 @@
         }
       }
     },
-    {
-      "name": "avg_amount_group_by_integer_esql_segment_partitioning",
-      "operation-type": "raw-request",
-      "method": "POST",
-      "path": "/_esql",
-      "body": {
-        "query" : "from nyc_taxis | stats avg(total_amount) by passenger_count | sort passenger_count",
-        "pragma" : {
-          "data_partitioning" : "segment"
-        }
-      }
-    },
     {
       "name": "avg_amount_group_by_integer_esql_doc_partitioning",
       "operation-type": "raw-request",
@@ -876,18 +847,6 @@
         }
       }
     },
-    {
-      "name": "avg_amount_group_by_keyword_esql_segment_partitioning",
-      "operation-type": "raw-request",
-      "method": "POST",
-      "path": "/_esql",
-      "body": {
-        "query" : "from nyc_taxis | stats avg(total_amount) by rate_code_id | sort rate_code_id",
-        "pragma" : {
-          "data_partitioning" : "segment"
-        }
-      }
-    },
     {
       "name": "avg_amount_group_by_keyword_esql_doc_partitioning",
       "operation-type": "raw-request",
@@ -944,18 +903,6 @@
         }
       }
     },
-    {
-      "name": "avg_passenger_count_filtered_esql_segment_partitioning",
-      "operation-type": "raw-request",
-      "method": "POST",
-      "path": "/_esql",
-      "body": {
-        "query" : "from nyc_taxis | where total_amount > 60 and rate_code_id==\"2\"| stats avg(passenger_count)",
-        "pragma" : {
-          "data_partitioning" : "segment"
-        }
-      }
-    },
     {
       "name": "avg_passenger_count_filtered_esql_doc_partitioning",
       "operation-type": "raw-request",
@@ -993,19 +940,6 @@
         }
       }
     },
-    {
-      "name": "sort_by_ts_esql_segment_partitioning",
-      "operation-type": "raw-request",
-      "method": "POST",
-      "path": "/_esql",
-      "request-timeout": 120,
-      "body": {
-        "query" : "from nyc_taxis | sort pickup_datetime desc | project pickup_datetime, dropoff_datetime, trip_distance| limit 1000",
-        "pragma" : {
-          "data_partitioning" : "segment"
-        }
-      }
-    },
     {
       "name": "sort_by_ts_esql_doc_partitioning",
       "operation-type": "raw-request",