[GitHub] [doris] HappenLee commented on a diff in pull request #12586: [Opt](vectorized) Use new way to do hash shffle to speed up query

[GitBox <gi...@apache.org>]

HappenLee commented on code in PR #12586:
URL: https://github.com/apache/doris/pull/12586#discussion_r970786434


##########
gensrc/thrift/PaloInternalService.thrift:
##########
@@ -175,6 +175,8 @@ struct TQueryOptions {
 
   // For debug purpose, skip delete predicates when reading data
   49: optional bool skip_delete_predicate = false
+
+  50: optional bool enable_new_shuffle_hash_method = true

Review Comment:
   I just want keep the name if next time we need change shuffe hash method, still can use the value 



##########
be/src/vec/sink/vdata_stream_sender.h:
##########
@@ -316,14 +318,13 @@ class VDataStreamSender::Channel {
     bool _enable_local_exchange = false;
 };
 
-template <typename Channels, typename HashVals>
+template <typename Channels>
 Status VDataStreamSender::channel_add_rows(Channels& channels, int num_channels,
-                                           const HashVals& hash_vals, int rows, Block* block) {
+                                           uint64_t* __restrict hash_vals, int rows, Block* block) {
     std::vector<int> channel2rows[num_channels];
 
     for (int i = 0; i < rows; i++) {
-        auto cid = hash_vals[i] % num_channels;
-        channel2rows[cid].emplace_back(i);
+        channel2rows[hash_vals[i]].emplace_back(i);

Review Comment:
   OK



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org