[native] Add config for cpp cache default behavior

In Presto Java, cache is disabled by default and only enabled when the config `node_preference` is set to `SOFT_AFFINITY`, while in CPP, cache is enabled by default and only disabled when the config node_preference is set to "NO_PREFERENCE". This change adds a new config to support the alignment of the default cache behavior.
unix280 · Nov 20, 2024 · ec60c13 · ec60c13
1 parent 72f3cef
commit ec60c13
Show file tree

Hide file tree

Showing 4 changed files with 41 additions and 6 deletions.
diff --git a/presto-docs/src/main/sphinx/presto_cpp/features.rst b/presto-docs/src/main/sphinx/presto_cpp/features.rst
@@ -184,6 +184,17 @@ disabled if ``connector.num-io-threads-hw-multiplier`` is set to zero.
 
 Whether async data cache is enabled.
 
+``query-data-cache-enabled-default``
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+* **Type** ``bool``
+* **Default value:** ``true``
+
+If ``true``, SSD cache is enabled by default and is disabled only if
+``node_selection_strategy`` is present and set to ``NO_PREFERENCE``.
+Otherwise, SSD cache is disabled by default and is enabled if
+``node_selection_strategy`` is present and set to ``SOFT_AFFINITY``.
+
 ``async-cache-ssd-gb``
 ^^^^^^^^^^^^^^^^^^^^^^
 

diff --git a/presto-native-execution/presto_cpp/main/QueryContextManager.cpp b/presto-native-execution/presto_cpp/main/QueryContextManager.cpp
@@ -98,15 +98,26 @@ void updateVeloxConnectorConfigs(
     std::unordered_map<
         std::string,
         std::unordered_map<std::string, std::string>>& connectorConfigStrings) {
+  const auto& systemConfig = SystemConfig::instance();
   for (auto& entry : connectorConfigStrings) {
     auto& connectorConfig = entry.second;
-
-    // Do not retain cache if `node_selection_strategy` is explicitly set to
-    // `NO_PREFERENCE`.
+    // If queryDataCacheEnabledDefault is true, when `node_selection_strategy`
+    // is
+    //       not set                             retain cache
+    //       SOFT_AFFINITY                       retain cache
+    //       NO_PREFERENCE                       do not retain cache
+    // If queryDataCacheEnabledDefault is false, when `node_selection_strategy`
+    // is
+    //       not set                             do not retain cache
+    //       SOFT_AFFINITY                       retain cache
+    //       NO_PREFERENCE                       do not retain cache
+    connectorConfig.emplace(
+        connector::hive::HiveConfig::kCacheNoRetentionSession,
+        systemConfig->queryDataCacheEnabledDefault() ? "false" : "true");
     auto it = connectorConfig.find("node_selection_strategy");
-    if (it != connectorConfig.end() && it->second == "NO_PREFERENCE") {
-      connectorConfig.emplace(
-          connector::hive::HiveConfig::kCacheNoRetentionSession, "true");
+    if (it != connectorConfig.end()) {
+      connectorConfig[connector::hive::HiveConfig::kCacheNoRetentionSession] =
+          it->second == "SOFT_AFFINITY" ? "false" : "true";
     }
   }
 }

diff --git a/presto-native-execution/presto_cpp/main/common/Configs.cpp b/presto-native-execution/presto_cpp/main/common/Configs.cpp
@@ -178,6 +178,7 @@ SystemConfig::SystemConfig() {
           BOOL_PROP(kNativeSidecar, false),
           BOOL_PROP(kAsyncDataCacheEnabled, true),
           NUM_PROP(kAsyncCacheSsdGb, 0),
+          BOOL_PROP(kQueryDataCacheEnabledDefault, true),
           NUM_PROP(kAsyncCacheSsdCheckpointGb, 0),
           STR_PROP(kAsyncCacheSsdPath, "/mnt/flash/async_cache."),
           NUM_PROP(kAsyncCacheMaxSsdWriteRatio, 0.7),
@@ -464,6 +465,10 @@ bool SystemConfig::asyncDataCacheEnabled() const {
   return optionalProperty<bool>(kAsyncDataCacheEnabled).value();
 }
 
+bool SystemConfig::queryDataCacheEnabledDefault() const {
+  return optionalProperty<bool>(kQueryDataCacheEnabledDefault).value();
+}
+
 uint64_t SystemConfig::asyncCacheSsdCheckpointGb() const {
   return optionalProperty<uint64_t>(kAsyncCacheSsdCheckpointGb).value();
 }

diff --git a/presto-native-execution/presto_cpp/main/common/Configs.h b/presto-native-execution/presto_cpp/main/common/Configs.h
@@ -313,6 +313,12 @@ class SystemConfig : public ConfigBase {
 
   static constexpr std::string_view kAsyncDataCacheEnabled{
       "async-data-cache-enabled"};
+  /// If true, SSD cache is enabled by default and is disabled only if
+  /// `node_selection_strategy` is present and set to `NO_PREFERENCE`.
+  /// Otherwise, SSD cache is disabled by default and is enabled if
+  /// `node_selection_strategy` is present and set to `SOFT_AFFINITY`.
+  static constexpr std::string_view kQueryDataCacheEnabledDefault{
+      "query-data-cache-enabled-default"};
   static constexpr std::string_view kAsyncCacheSsdGb{"async-cache-ssd-gb"};
   static constexpr std::string_view kAsyncCacheSsdCheckpointGb{
       "async-cache-ssd-checkpoint-gb"};
@@ -752,6 +758,8 @@ class SystemConfig : public ConfigBase {
 
   bool asyncDataCacheEnabled() const;
 
+  bool queryDataCacheEnabledDefault() const;
+
   uint64_t asyncCacheSsdGb() const;
 
   uint64_t asyncCacheSsdCheckpointGb() const;