title: "Materialize v0.39" date: 2023-01-11 released: true
Add mz_internal.mz_source_statistics
to the system catalog. This table
contains statistics for each process of each source in the system, like the
number of messages and bytes received from the upstream external system.
Add mz_internal.mz_object_dependencies
to the system catalog. This table
describes the dependency structure between all objects in Materialize. As an
example, you can now get an overview of the relationship between user-defined
objects using:
SELECT
object_id,
o.name,
o.type,
referenced_object_id,
ro.name,
ro.type
FROM mz_internal.mz_object_dependencies
JOIN mz_objects o ON object_id = o.id
JOIN mz_objects ro ON referenced_object_id = ro.id
WHERE o.id LIKE 'u%' AND ro.id NOT LIKE 's%'
ORDER BY o.name DESC, ro.name ASC;
It's important to note that these tables are part of an unstable interface of
Materialize (mz_internal
), which means that their values may change at any
time, and you should not rely on them for tasks like capacity planning for
the time being.
Add an mz_version
system configuration parameter, which reports the
Materialize version information. The value of this parameter is the same as
the value returned by the existing mz_version()
function, but the parameter
form can be more convenient for downstream applications.
SHOW mz_version;
mz_version
---------------------
v0.39.2 (e6af8921b)
Automatically create a linked cluster associated with each source and sink.
The mappings between sources/sinks and their respective linked cluster are
exposed in the mz_internal.mz_cluster_links
system catalog table.
The concept of a linked cluster is not user-facing, and is intentionally
undocumented. Linked clusters are meant to preserve the soon-to-be legacy
interface for sizing sources and sinks, where a SIZE
parameter is specified
on the source/sink rather than the cluster replica.
Add the IDLE ARRANGEMENT MERGE EFFORT
advanced option to CREATE CLUSTER
REPLICA
, which enables configuring the amount of effort a replica exerts on
compacting arrangements during idle periods.
Private preview. Support bearer token authentication in the WebSocket API endpoint, which supports interactive SQL queries over WebSockets.