Will Berkeley created KUDU-2731:
-----------------------------------
Summary: Getting column schema information from KuduSchema
requires copying a KuduColumnSchema object
Key: KUDU-2731
URL: https://issues.apache.org/jira/browse/KUDU-2731
Project: Kudu
Issue Type: Improvement
Affects Versions: 1.9.0
Reporter: Will Berkeley
I'm looking at a CPU profile of Impala inserting into Kudu.
{{KuduTableSink::Send}} has code that schematically does the following:
{noformat}
for each row in the batch
for each column
if (schema.Column(col_idx).isNullable()) {
write->mutable_row()->SetNull(col);
}
}
}
{noformat}
See
[kudu-table-sink.cc|https://github.com/apache/impala/blob/branch-3.1.0/be/src/exec/kudu-table-sink.cc#L236].
However, {{KuduSchema::Column}} copies the column schema and returns it by
value, so the if statement constructs and destroys a column schema object just
to check if the column is nullable.
This is by far the biggest user of CPU in the Impala process (35% or so). The
workload might be I/O bound writing to Kudu anyway, though. Nevertheless, we
should provide a way to avoid this copying in the API, either by adding a
method like
{noformat}
class KuduSchema {
const KuduColumnSchema& get_column(int idx);
}
{noformat}
or a method like
{noformat}
class KuduSchema {
bool is_column_nullable(int idx);
}
{noformat}
The former is the most flexible while the latter frees the client from worrying
about holding the ref longer than the KuduColumnSchema object lives. We might
need to add a number of methods similar to the latter method to cover other
potentially useful things like checking encoding, type, etc.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
