diff options
| author | Kevin Grittner <kgrittn@postgresql.org> | 2014-10-13 10:16:36 -0500 |
|---|---|---|
| committer | Kevin Grittner <kgrittn@postgresql.org> | 2014-10-13 10:16:36 -0500 |
| commit | 30d7ae3c76d2de144232ae6ab328ca86b70e72c3 (patch) | |
| tree | 1ef25acf6cbb5843eff05a82e3282d4c314d7bea /src/include/executor | |
| parent | 494affbd900d1c90de17414a575af1a085c3e37a (diff) | |
| download | postgresql-30d7ae3c76d2de144232ae6ab328ca86b70e72c3.tar.gz postgresql-30d7ae3c76d2de144232ae6ab328ca86b70e72c3.zip | |
Increase number of hash join buckets for underestimate.
If we expect batching at the very beginning, we size nbuckets for
"full work_mem" (see how many tuples we can get into work_mem,
while not breaking NTUP_PER_BUCKET threshold).
If we expect to be fine without batching, we start with the 'right'
nbuckets and track the optimal nbuckets as we go (without actually
resizing the hash table). Once we hit work_mem (considering the
optimal nbuckets value), we keep the value.
At the end of the first batch, we check whether (nbuckets !=
nbuckets_optimal) and resize the hash table if needed. Also, we
keep this value for all batches (it's OK because it assumes full
work_mem, and it makes the batchno evaluation trivial). So the
resize happens only once.
There could be cases where it would improve performance to allow
the NTUP_PER_BUCKET threshold to be exceeded to keep everything in
one batch rather than spilling to a second batch, but attempts to
generate such a case have so far been unsuccessful; that issue may
be addressed with a follow-on patch after further investigation.
Tomas Vondra with minor format and comment cleanup by me
Reviewed by Robert Haas, Heikki Linnakangas, and Kevin Grittner
Diffstat (limited to 'src/include/executor')
| -rw-r--r-- | src/include/executor/hashjoin.h | 5 |
1 files changed, 5 insertions, 0 deletions
diff --git a/src/include/executor/hashjoin.h b/src/include/executor/hashjoin.h index c9e61dfa39c..0e1e0cd5f0f 100644 --- a/src/include/executor/hashjoin.h +++ b/src/include/executor/hashjoin.h @@ -127,6 +127,10 @@ typedef struct HashJoinTableData int nbuckets; /* # buckets in the in-memory hash table */ int log2_nbuckets; /* its log2 (nbuckets must be a power of 2) */ + int nbuckets_original; /* # buckets when starting the first hash */ + int nbuckets_optimal; /* optimal # buckets (per batch) */ + int log2_nbuckets_optimal; /* same as log2_nbuckets optimal */ + /* buckets[i] is head of list of tuples in i'th in-memory bucket */ struct HashJoinTupleData **buckets; /* buckets array is per-batch storage, as are all the tuples */ @@ -148,6 +152,7 @@ typedef struct HashJoinTableData bool growEnabled; /* flag to shut off nbatch increases */ double totalTuples; /* # tuples obtained from inner plan */ + double skewTuples; /* # tuples inserted into skew tuples */ /* * These arrays are allocated for the life of the hash join, but only if |