The worker pool (used by executeBlocking) is a resource that can limit concurrency in your application, from the concurrency perspective you can look at Little's law for a simple explanation. In addition the conversion also consume CPU.
Is there a cost of using execute blocking for event handling? I have a service converting Json and xml at high speed which sometimes triggers a warning for execution time. I have to handle around 6 to 20k events per sec #vertx @vertx_project
-
-
-
Thanks I can work with this!
Kraj razgovora
Novi razgovor -
-
-
If the problem is at parsing of json, you can have a look at the asynchronous json parser: https://vertx.io/docs/vertx-core/java/#_json_parser …
-
Thanks
@pml0pes! I’ll look into it to see if this helps. Do also have xml messages coming in though. I still have to profile the application but I have a hunch that the xml messages take a lot of time. Luckily 20% of events are xml the rest is json - Još 1 odgovor
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.