[IMPALA-6461] *DataStreamSender::Channel::AddRow needs some micro-optimizations to remove per row function call and data dependency

• Type: Improvement
• Status: Resolved
• Priority: Minor
• Resolution: Fixed
• Fix Version/s: Impala 2.11.0
• Component/s: Distributed Exec
• Epic Color: ghx-label-4
• Assignee: Mostafa Mokhtar
• Reporter: Mostafa Mokhtar
• Created: 31/Jan/18 00:26
• Updated: 22/Feb/18 19:42
• Resolved: 22/Feb/18 19:42

Description

While analyzing performance of partition exchange operator I noticed that there is dependency and a function call per row in the hot path.

// hash-partition batch's rows across channels
 // TODO: encapsulate this in an Expr as we've done for Kudu above and remove this case
 // once we have codegen here.
 int num_channels = channels_.size();
 for (int i = 0; i < batch->num_rows(); ++i) {
 TupleRow* row = batch->GetRow(i);
 uint64_t hash_val = EXCHANGE_HASH_SEED;
 for (int j = 0; j < partition_exprs_.size(); ++j) {
 ScalarExprEvaluator* eval = partition_expr_evals_[j];
 void* partition_val = eval->GetValue(row);
 // We can't use the crc hash function here because it does not result in
 // uncorrelated hashes with different seeds. Instead we use FastHash.
 // TODO: fix crc hash/GetHashValue()
 DCHECK(&(eval->root()) == partition_exprs_[j]);
 hash_val = RawValue::GetHashValueFastHash(
 partition_val, partition_exprs_[j]->type(), hash_val);
 }
 RETURN_IF_ERROR(channels_[hash_val % num_channels]->AddRow(row));
 }

Force inlining DataStreamSender::Channel::AddRow and breaking up the loop improves partition exchange performance by 5%

Code-generation of the hash computation IMPALA-5168 should give another 10% speedup.