Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Hey There

Hey there 🙂 Still my journey to deploy the aws-autoscaler with spot instances, I have another question:
I would like to limit the amount of time spent setting up the experiments in the spot instances, in particular, I would like to make the data used for training already available and not have to download it from s3 each time a new spot instance is started. I want to create an EBS disk with the data used for training, create a snapshot of it, and then make the autoscaler attach a copy of this EBS snapshot to every spot instance it spins up. How can I do that with the aws-autoscaler?

Posted 2 years ago
Votes Newest

Answers 2

Alright, thanks SuccessfulKoala55 !

Posted 2 years ago

Hi JitteryCoyote63 ,
You can do this by modifying the startup_bash_script and launch_specification used in AwsAutoScaler.spin_up_worker() - for that, you'll probably need to define your own class AwsAutoScalerExt(AwsAutoScaler) and make sure it's used in the example code.

Posted 2 years ago
2 Answers
2 years ago
one year ago
Similar posts