Hi all,
I have been looking into using Flink in batch mode to process Iceberg
tables. I noticed that the performance for queries in Flink's batch mode is
quite slow, especially when compared to Spark. I'm wondering if there are
any configurations that I'm missing to get better performance out of
r
|Order_1 |
++--++
Output:
6> +I[Order_1, 1]
6> -U[Order_1, 1]
Here is the link to a github repository that contains an example. (
https://github.com/charles-tan/flink-upsert-changelog-bug)
Thanks,
Charles
vepoint. (https://github.com/charles-tan/flink-state-processor-example)
Thanks,
Charles
Hi all,
I’ve tried a simple Flink application which uses FlinkKinesisConsumer. I
noticed that when trying to consume from Kinesalite using the
FlinkKinesisConsumer with EFO enabled, I run into SSL handshake errors.
This is despite disabling certificate validation. Has anybody successfully
tested F
context on interval
joins and can explain cleanUpTime and minCleanUpInterval.
Thanks,
Charles
On Tue, Mar 14, 2023 at 1:44 PM Charles Tan wrote:
> Hi everyone,
>
> I have been playing around with Flink SQL’s interval joins and noticed
> that some outputs from unmatched LEFT or FULL joins a
Hi everyone,
I have been playing around with Flink SQL’s interval joins and noticed that
some outputs from unmatched LEFT or FULL joins are arriving much later than
I expected. Take the following query for example:
SELECT * FROM orders o LEFT JOIN shipments s
ON (o.orderID = s.orderID) AND o.rowti
27;ve updated both tickets with comments and new code
snippets. Is there another way to load UDFs in Flink 1.16?
Code examples for reference:
https://github.com/charles-tan/udfs-flink-1.16/blob/main/src/test/java/com/example/UDFTest.java
Thanks,
Charles
On Fri, Nov 4, 2022 at 10:51 AM Alexander Fe
