AncientSoul

joined 3 weeks ago
[–] AncientSoul@reddthat.com 2 points 12 hours ago

Mine is close to ten years old and still going strong. We have one at work as well. It is heavily used to feed two Wilfa Svart coffee makers multiple times a day without issues.

[–] AncientSoul@reddthat.com 7 points 16 hours ago (3 children)

The Wilfa Svart grinders are really good. Wilfa is a Norwegian company and the Svart line of coffee making appliances is developed in cooperation with former world champion barista Tim Wendelboe.

[–] AncientSoul@reddthat.com 3 points 1 day ago

Most of the providers are targeting web hosting and general fairly constant load. We have bursts load and use queue based scaling in AWS. Does anyone know of any service that can do that?

[–] AncientSoul@reddthat.com 1 points 3 days ago

Yeah, your nick gives you away, it I didn’t think about that when writing that 😁

[–] AncientSoul@reddthat.com 1 points 3 days ago (2 children)

Norgesgruppen lista SPAR as one of their brands/concepts. Unfortunately, their Norwegian / English toggle doesn’t work, but this is the Norwegian version: https://www.norgesgruppen.no/om-oss/om-oss-hjem/

[–] AncientSoul@reddthat.com 2 points 3 days ago (4 children)

We do, but in an odd way. One of the Norwegian chains use Spar as a brand for some of their stores.

[–] AncientSoul@reddthat.com 12 points 4 days ago (8 children)

None of these chains are present in Norway. Lidl tried years ago and gave up. We only have some Norwegian chains here.

[–] AncientSoul@reddthat.com 1 points 1 week ago

Agreed. At least we have ARM who might start to produce their own chips. We know how to design them, just not capable of actually producing them.

 

DataCrunch is based in Finland and provides long time rental of bare metal servers, short time use of instances similar to EC2 and serverless container hosting. The latter is particularly interesting since they come with autoscaling and queue support out of the box. We have been using Fargate, but then you can’t go completely serverless with GPUs and the queue is a separate entity. We deployed our first model using a vLLM docker image in days without having used the system before. We will probably moving existing model hosting from AWS to DataCrunch as well.