Dynamically adjusting performance of materialized view maintenance

A method, comprising:

• monitoring, by one or more computing devices implementing a materialized view management platform, respective performance of one or more computing resources allocated to maintain a materialized view when the materialized view is created by the materialized view management platform to:
• obtain changes made via respective interfaces to one or more of a plurality of different data sources used to create a materialized view stored in a target data store, wherein the one or more computing resources of the materialized view management platform are implemented separately from different respective storage systems implementing the plurality of different data sources and a storage system implementing the target data store;
• determine updates to be made to the materialized view according to respective operation assignments, wherein the respective operation assignments select between local performance at the allocated computing resources or remote performance at one of the one or more data sources, wherein the respective operation assignments are determined based, at least in part, on computation capabilities of the one or more data sources;
• and
• performing target-specific update translation that performs at least one data type conversion or operation conversion in addition to generating request parameters according to an interface of the target data store to send one or more requests with the generated request parameters to make the determined updates to the materialized view stored in the target data store, wherein the interface of the target data store supports one or more different features than the respective interfaces of the one or more data sources;
• detecting an event to adjust the performance of updating the materialized view based on the monitoring;
• in response to detecting the event:
• scaling the allocated computing resources to adjust the performance of updating the materialized view based on the detected event, wherein the scaling of the allocated computing resources causes an adjustment to subsequent operations to perform at least one of:
• obtaining further changes for the materialized view from the different data sources;
• determining further updates to make to the materialized view according to further respective operation assignments, wherein the further respective operation assignments select between local performance at the scaled computing resources or remote performance at one of the one or more data sources, wherein the further respective operation assignments are determined based, at least in part, on the computation capabilities of the one or more data sources;
• or
• performing target-specific update translation to generate further request parameters according to the interface of the target data store to send one or more further requests with the generated request parameters to make the further updates to the materialized view in the target data store.

One or more non-transitory, computer-readable storage media, storing program instructions that when executed on or across one or more computing devices cause the one or more computing devices to implement a materialized view management platform that implements:

• monitoring, by one or more computing devices implementing a materialized view management platform, respective performance of one or more computing resources allocated to maintain a materialized view when the materialized view is created by the materialized view management platform to:
• obtain changes made via respective interfaces to a plurality of different data sources used to create a materialized view stored in a target data store, wherein the one or more computing resources of the materialized view management platform are implemented separately from different respective storage systems implementing the plurality of different data sources and a storage system implementing the target data store;
• determine updates to be made to the materialized view according to respective operation assignments, wherein the respective operation assignments select between local performance at the allocated computing resources or remote performance at one of the one or more data sources, wherein the respective operation assignments are determined based, at least in part, on computation capabilities of the one or more data sources;
• and
• performing target-specific update translation that performs at least one data type conversion or operation conversion in addition to generating request parameters according to an interface of the target data store to send one or more requests with the generated request parameters to make the determined updates to the materialized view stored in the target data store, wherein the interface of the target data store supports one or more different features than the respective interfaces of the plurality of different data sources;
• detecting an event to adjust the performance of updating the materialized view based on the monitoring;
• in response to detecting the event:
• scaling the allocated computing resources to adjust the performance of updating the materialized view based on the detected event, wherein the scaling of the allocated computing resources causes an adjustment to subsequent operations to perform at least one of:
• obtaining further changes for the materialized view from the different data sources;
• determining further updates to make to the materialized view according to further respective operation assignments, wherein the further respective operation assignments select between local performance at the scaled computing resources or remote performance at one of the one or more data sources, wherein the further respective operation assignments are determined based, at least in part, on the computation capabilities of the one or more data sources;
• or
• performing target-specific update translation to generate further request parameters according to the interface of the target data store to send one or more further requests with the generated request parameters to make the further updates to the materialized view in the target data store.