-
Hello, currently, I am running Nextflow on a Kubernetes cluster and trying to improve its performance. While going through the logs, I noticed that within these steps, it took nearly 10 minutes, whereas running it in a local environment took only a few seconds. ![]() The nextflow version is 23.04.3. Can anyone help me optimize this or provide any reasons behind this? Thanks a lot. |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 3 replies
-
Finally I have been able to solve this problem. The reason for this delay is that: First, I am using the nextflow docker image from docker hub to start the pipeline. Apparently that before nextflow parsing script and stuff, it downloads some dependencies (maybe related with nf-core or my pipeline itself). And also because of my k8s cluster was placed in Aliyun (Alibaba cloud) Container Service, the internet from the pod to outside was very slow, that caused this delay. |
Beta Was this translation helpful? Give feedback.
Finally I have been able to solve this problem. The reason for this delay is that: First, I am using the nextflow docker image from docker hub to start the pipeline. Apparently that before nextflow parsing script and stuff, it downloads some dependencies (maybe related with nf-core or my pipeline itself). And also because of my k8s cluster was placed in Aliyun (Alibaba cloud) Container Service, the internet from the pod to outside was very slow, that caused this delay.
So my solution for this was creating a new image from the nextflow docker image and then running the pipeline first so it can download all the stuff. After that pushing the image to any Container Registry and using that ima…