New MySQL HeatWave capabilities released in first half of 2022
MySQL HeatWave has introduced several new capabilities in the first half of 2022. This includes new functionalities, features to improve operational efficiency and further improving the query performance. MySQL HeatWave is supported in all commercial OCI regions and several new regions were added in the last few months.
New Functionality
HeatWave Machine Learning
With HeatWave ML, you can train a model, generate inferences and offer machine learning explanations, without extracting data out of the MySQL database. HeatWave ML fully automates the training process and creates a model with the best algorithm, optimal features, and the optimal hyper-parameters for a given data set and a specified task. The performance of HeatWave ML training is 25 times faster than other service, so you can now retrain models more frequently and keep up with changes to data. This keeps the models up-to-date which improves the accuracy of predictions. This enables you to accelerate your ML initiatives, improve security of your data and application, and reduce costs. Explore HeatWave ML here.
Real-time Elasticity: Resize HeatWave cluster with no downtime
Now you can resize your HeatWave cluster to any cluster size without any downtime or read-only time. The resizing operation takes only a few minutes, during which time HeatWave remains online, available for all operations. Once resized, data is downloaded from HeatWave storage, automatically repartitioned among all available cluster nodes, and becomes immediately available for queries. For more information, visit here.
Pause and resume HeatWave cluster
With efficient data reload from HeatWave storage, you can now pause and resume HeatWave cluster to reduce cost. When you pause the cluster, the system instantaneously stops the cluster. Upon resume, the system reloads the data, metadata and MySQL Autopilot statistics in the HeatWave memory at near network bandwidth. Since the MySQL Autopilot statistics and metadata are also saved and restored, you can pause the cluster without losing the workload history. For more details, click here.
Operational Efficiency
Data Compression: 2X more data processed per HeatWave node
HeatWave has introduced data compression which allows each HeatWave node to process up to 2X more data without any degradation in price performance for queries. This reduces the number of HeatWave nodes needed to process your workload and cut your cost by up to 50%. For more information, click here.
Eliminate minimum HeatWave cluster size
HeatWave has eliminated the minimum cluster size requirement. This reduces the entry cost of a HeatWave cluster. You can now use HeatWave for smaller workloads with minimum cost.
Efficient reload of data in HeatWave cluster
HeatWave maintains the copy of the in-memory representation of data, metadata and MySQL Autopilot statistics in the HeatWave storage (which resides in the object store) and continuously updates it. When HeatWave cluster is restarted (planned or unplanned), data is loaded in parallel from HeatWave storage to HeatWave node. The time it takes to load only depends on the amount of data per node, it is independent of the overall data size. This results in significant improvement in reload performance over reloading from MySQL InnoDB. (100x improvement for 10TB data). Learn more here.
Auto recovery on stale dictionary encoded table
For some queries, performance can be improved by encoding string columns (dictionary encoding) used in the query. HeatWave can support more than 4 billion unique values in the string column with dictionary encoding. To save memory space, HeatWave does not allocate memory for all 4 billion values. As your data grows and number of unique values increases, HeatWave may run out of space for the pre-allocated memory. In this case, HeatWave will automatically reload the table to increase the amount of pre-allocated memory, allowing you to continue accelerate queries in HeatWave without user intervention. For more information, click here
Query Performance
Support for Views
MySQL HeatWave introduced support for accelerating queries with views. Queries executed on views have the same offload prerequisites as queries executed on the base tables—for example, the table needs to be loaded into HeatWave.
Native IN-clause Support:
HeatWave now supports native execution of IN-clause. During execution the IN-clause list elements are stored in a cache efficient data structure and the column is streamed through by each core in the HeatWave node. This technique provides up to 100x acceleration for queries with large number of elements in the IN-clause list.
More columns and wider columns
The supported number of columns in base relation loaded into HeatWave has been increased from 473 to 1,017 columns, which is also the InnoDB column limit. The VARCHAR/TEXT column width limit has been increased from 8KB to 64KB.
Global Expansion
New regions: Johannesburg in South Africa, Paris in France
MySQL HeatWave has expanded its availability to Johannesburg in South Africa and Paris in France. With these regions, MySQL HeatWave now supports 31 global regions on OCI. For more details, click here.
Additional Resources
MySQL HeatWave Product Manager
Next Post