Marco Slot
Marco Slot
Users currently have 2 sets of size functions: - pg_*_size - citus_*_size The PostgreSQL functions give incorrect sizes for distributed tables & indexes on distributed tables. The Citus functions give...
DESCRIPTION: Add citus.distributed_data_dump to only return results from local shards For CDC purposes, it is useful to have a way to get only local data from distributed tables from each...
The --bgworker option seems useful, but it spits out quite outdated code. ``` $ cargo pgx new mybgw --bgworker $ cd mybgw $ cargo pgx run pg15 error[E0599]: no function...
It would be useful if `pgcopydb list progress` would give more detailed information such as the number of bytes copied and maybe information from `pg_stat_progress_create_index`.
For migration scenarios where the source is at PostgreSQL 10 or newer, it would be useful if pgcopydb follow mode could use logical replication (CREATE PUBLICATION & CREATE SUBSCRIPTION). A...
pgcopydb installs into the PostgreSQL installation directory. It seems possible to also provide PostgreSQL extension (.control + .sql + .so) with functions to run the pgcopydb commands from SQL. In...
I seem to repeatedly get the following assert failure when running tests on PG14: ``` TRAP: FailedAssertion("flags & HASH_STRINGS", File: "dynahash.c", Line: 427, PID: 2936) postgres: marco contrib_regression [local] SELECT(ExceptionalCondition+0xa5)[0x55a1fcdaa8a9]...
We could make the change replication in Podyn callable as a Lambda function by implementing a `RequestHandler` per the example in: http://docs.aws.amazon.com/lambda/latest/dg/with-dynamodb-create-package.html#with-ddb-example-deployment-pkg-java Podyn is already packed in the right way...
Podyn can be set up on multiple nodes for scaling out or fault tolerance. However, this currently only supported in `jsonb` mode. In `columns` mode, nodes may independently decide to...
We often run HammerDB concurrently with the analytical queries from [CH-benCHmark](https://db.in.tum.de/research/projects/CHbenCHmark/) (mainly against PostgreSQL) by using a separate script. Would there be interest in adding a CH workload directly to...