Connectors Google Pub/Sub

Google Pub/Sub data integration with Lima, in both directions.

Pub/Sub is Google Cloud's messaging service. Lima pulls from subscriptions in micro-batches, acknowledges messages only after they are loaded, and publishes results to topics.

SOURCEGoogle Pub/SubSubscription pulls · PublishingLIMA BACKBONEExtractfull · incremental · streamingCheckevery row valid, or an exception with a…Transform and load150+ operators · Python · loads on the…TARGETSPlanning applicationsKinaxis · o9 · Blue Yonder · IBP · OMPWarehouses and lakehousesSnowflake · Databricks · Fabric · S3 …Reporting and partnersPower BI · Tableau · EDI · filesRESULTS RETURN TO GOOGLE PUB/SUB · EACH WRITE CONFIRMED
How Google Pub/Sub fits the six Backbone steps. It is a source and a target: data comes out through the methods below, and results go back in with confirmation.
2
ways to read or write it
this connector
4
patterns supported
this connector
3
ways to run: SaaS, private cloud, on-premise
product
22 days
from contract to production
showcase estate
How Lima reads and writes Google Pub/Sub

Method by method.

Each method has a job. The first load uses the bulk method; the daily run uses the incremental or change-capture method; write-back uses the interface the system itself uses.

MethodWhat it doesWhen it is used
Subscription pullsReads messages; acknowledgement after load.Inbound.
PublishingPublishes results.Outbound.
What it covers

Objects and patterns.

The objects are the tables, views or records Lima reaches through the methods above. The patterns say how a copy is made and kept current.

ObjectPlain meaning
Topics and subscriptionsmessage streams
PatternSupported
Full loadYes
IncrementalYes
Change capture—
Streamingmicro-batch intervals
Write-backmessages published
Before you start

Prerequisites, then four steps.

The prerequisites are what the team that runs Google Pub/Sub grants. The four steps are the same for every connector.

  • A service account with Subscriber and Publisher roles
Register the connectionEndpoint, credentials from the vault, and the method. The connection is tested and a failed test names the cause.
Choose the objects and the patternPick tables or objects from the discovered schema, and the pattern: full, incremental, change capture or streaming.
Run the first load with checks onEvery row is valid or becomes an exception with an owner. The run ends with a count and a control total per table.
Schedule it on the business calendarA run window with a deadline; pause, resume and replay if a check trips; the report after every run.
What happens to a bad record

Incoming data = valid records + exception records. A row that fails a check is kept, routed to the person who owns it with a deadline, and replayed once fixed. The load carries on with the valid rows and the report says exactly what was held back. How exceptions work →

Terms on this page

Plain meanings.

ConnectorA packaged way to read or write one kind of system, with its authentication, methods and prerequisites described.
Full, incremental, change capture, streamingFour ways to make and keep a copy: read everything; read what changed since last time; read changes from a log as they happen; consume events in small batches.
Write-backWriting a result into the system that owns it through its released interface; the system's response is the confirmation.
Control totalThe sum of a key amount column compared between source and target, along with the row count, at the end of a run.
Exception recordA row that failed a check: kept, routed to an owner, replayed once fixed.
Questions

Asked about this connector.

Does Lima install anything in Google Pub/Sub?

Lima connects from outside through the interfaces listed above with credentials your administrators grant. Where a vendor requires its own agent or add-on for a method, the prerequisites say so.

How is load on Google Pub/Sub kept in check?

Parallel reads are capped and the cap is agreed with the team that runs the system. Large reads run inside an agreed run window. Change capture keeps the daily volume small.

What do we see when a run finishes?

A report per run: rows in, rows out, control totals and variance per table, the exceptions held back and who owns them, and whether the deadline was met.

Can we start small?

Yes. Most estates start with one source, one target and one dataset, then add tables and patterns on the same connection.

Try Google Pub/Sub on your own estate.

A trial connects this system, runs a first load and ends with a reconciliation report. Your administrators see the prerequisites before anything is installed.