Re: [I] KEDA scaler `pending_jobs` metric appears insufficient for scaling due to rapid task assignment by scheduler [datafusion-ballista]
SNagarajan2243 commented on issue #1701: URL: https://github.com/apache/datafusion-ballista/issues/1701#issuecomment-4501050442 @metegenez Yeah, that would be very helpful. I’ve also noticed some issues related to IPv6 while deploying on EKS, though I haven’t had the chance to fully investigate them yet. For the autoscaling discussion above, I’ll be eagerly waiting for your insights and guidance. Appreciate the help! -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [I] KEDA scaler `pending_jobs` metric appears insufficient for scaling due to rapid task assignment by scheduler [datafusion-ballista]
metegenez commented on issue #1701: URL: https://github.com/apache/datafusion-ballista/issues/1701#issuecomment-4496317709 Since i am not fully onboarded on that part, I will read about it soon and try to help you. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [I] KEDA scaler `pending_jobs` metric appears insufficient for scaling due to rapid task assignment by scheduler [datafusion-ballista]
SNagarajan2243 commented on issue #1701: URL: https://github.com/apache/datafusion-ballista/issues/1701#issuecomment-4479238306 @martin-g @metegenez @milenkovicm Would appreciate any clarification on whether `pending_jobs` is the right metric for autoscaling, or if there’s a better metric for actual executor load - also, if others could help clarify the questions above, that would be greatly appreciated. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
Re: [I] KEDA scaler `pending_jobs` metric appears insufficient for scaling due to rapid task assignment by scheduler [datafusion-ballista]
SNagarajan2243 commented on issue #1701: URL: https://github.com/apache/datafusion-ballista/issues/1701#issuecomment-4460974420 @andygrove -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] - To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
