MySQL HeatWave Release Notes
MySQL HeatWave AutoML queries are now monitored and recorded in the
Performance Schema tables rpd_query_stats and
rpd_exec_stats. Where a single MySQL HeatWave AutoML
query contains a number of sub-queries, there is one record in
rpd_query_stats and multiple records in
rpd_exec_stats.
(WL #15243)
New functions have been added to MySQL HeatWave AutoML to help you manage models:
When you run the ML_TRAIN routine on a
training dataset, you can now specify a model handle to use
for the model instead of the generated one.
A new column notes has been added to the
MODEL_CATALOG table, which you can use to
record notes about the models in your model catalog.
The new column model_metadata in the
MODEL_CATALOG table records metadata for
models, such as the training score, training time, and
information about the training dataset. If an error occurs
during training or you cancel the training operation,
MySQL HeatWave AutoML records the error status in this column.
(WL #15243)
MySQL HeatWave AutoML now supports the upload of pre-trained models in ONNX
(Open Neural Network Exchange) format to the model catalog. You
can load them using the stored procedure
ML_MODEL_IMPORT that provides the conversion
required to store the model in a MySQL table.
(WL #15243)
A new stored procedure ML_EXPLAIN lets you
train a variety of model explainers and prediction explainers
for MySQL HeatWave AutoML, in addition to the default Permutation
Importance model and prediction explainers:
The Partial Dependence model explainer shows how changing the values of one or more columns will change the value that the model predicts.
The SHAP model explainer produces global feature importance values based on Shapley values.
The Fast SHAP model explainer is a subsampling version of the SHAP model explainer which usually has a faster runtime.
The Permutation Importance prediction explainer explains the prediction for a single row or table.
The SHAP prediction explainer uses feature importance values to explain the prediction for a single row or table.
When you use the ML_EXPLAIN_TABLE and
ML_EXPLAIN_ROW stored procedures to generate
explanations for a prediction, you can now use the SHAP
prediction explainer as an alternative to the default
Permutation Importance prediction explainer. SHAP produces
feature importance values (explanations) based on Shapley
values.
(WL #15243)
MySQL HeatWave AutoML now supports timeseries forecasting using the
existing stored procedures ML_TRAIN,
ML_PREDICT_TABLE, and
ML_SCORE. You can create a forecast for a
single column (a univariate endogenous variable) with a numeric
data type. The forecasting task is specified as a JSON object
when you call the ML_TRAIN stored procedure.
(WL #15243)
MySQL HeatWave uses dictionary encoding to compress string columns
(CHAR, VARCHAR, TEXT). These dictionaries are built for each
string column with the
RAPID_COLUMN=ENCODING=SORTED keyword. MySQL HeatWave
now supports 8.5 billion dictionary entries (up from 4 billion),
which means MySQL HeatWave can now encode string columns with number of
distinct value (NDV) up to 8.5 billion.
(WL #14742)
MySQL HeatWave now uses statistics based on minimum and maximum values to create zone maps for every primary key column. MySQL HeatWave then uses the zone maps for range and point queries to only scan data chunks that are relevant for the query, and accelerates these queries by an order of magnitude. This is particularly useful for improving range queries in OLAP and mixed workloads. (WL #14713)
A new MySQL optimizer is introduced for MySQL HeatWave to provide a holistic cost model across MySQL and MySQL HeatWave, create better query plans based on statistics used in MySQL Autopilot, reduce compilation time, eliminate the need of query hints for join order, and improve join query performance. With the new optimizer, MySQL HeatWave can now run all 22 TPC-H queries without straight join hints. Before 8.0.31, a straight join hint is needed for 10 out of 22 TPC-H to reach peak performance. (WL #14449)
DDL statements such as ALTER TABLE,
RENAME TABLE, and TRUNCATE
TABLE are now permitted on a table that has RAPID
defined as the secondary engine. If a DDL operation is
successfully carried out on a table that is loaded to a
MySQL HeatWave Cluster at the time, MySQL HeatWave automatically reloads the table
from MySQL InnoDB. Note that if the DDL operation makes the
table’s structure incompatible with MySQL HeatWave, the table is
unloaded from the MySQL HeatWave Cluster.
(WL #15129)