gunnarmorling

8 ms·

61.
by gunnarmorling 1y ago
Author here, thanks for reading and commenting! Indeed my conclusion is that "agent" means different things to different people. The idea for this post was to explore what's there already and what may be missing for using SQL
62.
by gunnarmorling 1y ago
> It is the only database/query engine that allows you to use the same SQL for both batch and streaming (with UDFs). Flink SQL also checks that box.
63.
by gunnarmorling 1y ago
Congrats on the launch! Could you clarify perhaps what Archil volumes actually are from a technical perspective? Like, is it EC2 instances with locally attached storage, your own bare metal machines in a CoLo space, something in between? It
64.
by gunnarmorling 1y ago
Funny timing, just took a fresh look at this topic in this new post earlier this week: https://www.morling.dev/blog/backfilling-postgres-toast-colu... .
65.
Backfilling Postgres TOAST Columns in Debezium Data Change Events(morling.dev)
2 points by gunnarmorling 1y ago|0 comments
66.
by gunnarmorling 1y ago
That's one way to look at this, yes. The difference being that keys actually have a meaning to clients (as providers of ordering and also a failure domain), whereas partitions in their current form don't.
67.
by gunnarmorling 1y ago
> why is there no in-memory kafka server that I can use for simple testing purposes. Take a look at Debezium's KafkaCluster, which is exactly that: https://github.com/debezium/debezium/blob/main/d
68.
by gunnarmorling 1y ago
For each of the items discussed I explicitly mention why they would be desirable to have. How is this engineering for the sake of engineering?
69.
by gunnarmorling 1y ago
It can increase latency (which can be somewhat mitigated though by having a write buffer e.g. on EBS volumes), but it substantially _reduces_ cost: all cross-AZ traffic (which is $$$) is handled by the object storage layer, where it doesn&#
70.
by gunnarmorling 2y ago
Nice post for us escalator nerds out here! Here's two more folks might find interesting: - An escalator with a varying degree of ascent (it gets flatter and flatter) in Hamburg's Elbphilharmonie: https://www.youtube.com
71.
by gunnarmorling 2y ago
Leyden's AOT support is not specifically geared towards containerized workloads, AFAICS. It's a general trend in the Java world though to bundle the JVM alongside applications, potentially creating a tailored JVM distribution with
72.
by gunnarmorling 2y ago
Unless you are on a realtime OS, you may see longer pauses from the OS due to scheduling other processes, so I think it's fair to consider the new low-latency collectors (Shenandoah, ZGC) pauseless from a practical point of view.
73.
by gunnarmorling 2y ago
You don't need Kafka in order to run Debezium. While that certainly is the most common deployment, you can use Debezium also with things like Kinesis, Google Cloud Pub/Sub, or even plain HTTP sinks, all via Debezium Server ( https
74.
Let's Take a Look at KIP-932: Queues for Kafka(morling.dev)
2 points by gunnarmorling 2y ago|0 comments
75.
Failover Replication Slots with Postgres 17(decodable.co)
1 points by gunnarmorling 2y ago|0 comments
76.
by gunnarmorling 2y ago
Very nice to see JReleaser here on HN! I'm very happy with it, using it for publishing releases of kcctl (a command line client for Kafka Connect written in Java, compiled and published as native binaries for Linux/MacOS/Win
77.
by gunnarmorling 2y ago
Thanks, Brian! Big kudos to you and Sonatype for the service you provide to the Java community.
78.
by gunnarmorling 2y ago
In the Java world, you need to prove ownership of a given namespace (group id), e.g. via a TXT record for that domain. Isn't there a similar concept for NPM? The package is named sn4k-s3c/call-home, how will a victim be tricked in
79.
by gunnarmorling 2y ago
> Luckily [the accelerator servo] is also the only actuator that doesn’t come with major safety implications This is giving me chills, what if that RC servo gets "stuck" at full throttle? I suppose the assumption is you could h
80.
Failover Replication Slots with Postgres 17(decodable.co)
1 points by gunnarmorling 2y ago|0 comments
81.
by gunnarmorling 2y ago
Concurrent garbage collectors of recent JVM versions (ZGC, Shenandoah) can give you sub-millisecond pause times, i.e. GC pauses are not really an issue any more for typical analytics query workloads. The price to pay is reduced throughput,
82.
by gunnarmorling 2y ago
Indeed it's interesting to see aggregation as part of this list. Usually, the split of what can be pushed down to the storage layer and what cannot, is between stateless operations (filtering, projection) and stateful operations (e.g.
83.
by gunnarmorling 2y ago
A few potential reasons for this design coming to mind: - Resource allocation; you might want to give just specific amount of memory, CPU, network I/O to specific modules of a system, which is not really feasible within a single JVM -
84.
Revisiting the Outbox Pattern(decodable.co)
2 points by gunnarmorling 2y ago|0 comments
85.
by gunnarmorling 2y ago
> ... at event F52A Sure, if the database supports this sort of temporal query, then you're good with such id-only events. But that's not exactly the default for most databases / data models.
86.
by gunnarmorling 2y ago
Nice one. I wrote about this a while ago, from a slightly different perspective, focusing on data change events [1]. Making a similar differentiation there between id-only events (which you describe as triggers of action; from a data change
87.
by gunnarmorling 2y ago
It's not about mutability of events, but about mutating the underlying data itself. If the event only says "customer 123 has been updated", and a consumer of that event goes back to the source of the event to query the full s
88.
by gunnarmorling 2y ago
Thanks for sharing! I don't think people did exactly that, but most indeed did leverage the fact that values only ranged from -99.9 to 99.9 with exactly one fractional digit and handled them as integers (avoiding FP maths) up until the
89.
by gunnarmorling 2y ago
It's not that much, but it's also not nothing. The input file of the challenge was very narrow, just a relatively short text column and a numeric value. It still is >13 GB uncompressed; an actual table with a few more columns w
90.
by gunnarmorling 2y ago
Presenter of that talk here, very cool to see it being shared here. Running 1BRC was immensely fun, I learned a ton from it. Had you told me before how far the community would be able to push this, I'd have not believed you. One main t

More ›