Is it possible to contribute resources for Work Servers?

[Jassu]

Hi,

We're a registered non-profit association of computer hobbyists from Finland. As we have some spare resources available, would it be plausible to donate some VM capacity to ease out the overload situation with Work Servers the project is currently facing?

For us it would be easy to contribute at least 40 TB of fast storage, tens of cores of VCPU resources and up to 1 Gbps of continuous internet traffic. I can dig up more details regarding our readily available resources, if deploying new Work Servers would be feasible for you. All our infra is in a proper datacenter, located in Helsinki, Finland and connected redundantly through 10 Gbps transits. We are also prepared to support the project for at least the duration of the COVID-19 crisis, however long that might take.

[bruce]

I don't know, but I'll find somebody who does.

It would have to be integrated into a fairly complex system of other FAH servers and that's likely to be a problem to set up, at least initially.

[Jassu]

Thank you!

[Jassu]

I checked our available spare capacity. Let's just say, we should have enough spare to contribute properly.

There are way over 100 CPU cores sitting idle and a just a tad below 1 TB of RAM free. For us, it doesn't incur additional expenses to provide bandwidth up to 1 Gbps, but if there is need for more, we can have discussion if our transit can contribute to the COVID battle as well by waving any additional costs. As for storage, we can easily provide the 40 TB I mentioned earlier, but I believe there are more to spare if needed.

Our admins can contribute to the maintenance of the FAH system all the way to service level, or we can just provide you either pre-installed VMs with a distro of your choice or boot up VM images provided. If it's a highly custom setup, I would suggest going through the installation and maintenance tasks with our admins so that we can keep the services up & running with minimal effort from FAH staff.

The association already provides many publicly beneficial services in Finland and provides resources to other local non-profits also. The battle against COVID being the biggest disruption to all societies as of now, we see that joining the effort is directly in line with our purpose as a non-profit. Helping out on the work server side would give bigger benefits than providing just the CPU resources for already huge pool of computing capacity.

I hope this helps to catch a picture where we stand in both resources and effort we can put in.

Thank you.

[anandhanju]

Regardless of how it pans out, thanks Jassu. That is a very generous and thoughtful offer.

[jonkloske]

I'm also keen to know the answer to this. We run a fairly chunky Triton/SmartOS cluster at work which I think is 10Gbps, 960 Xeon cores running at 1.80GHz, 17TB of ram, 18TB disk.. not sure how much we could spare, but keep us posted.

[Neil-B]

iirc WS need large (50-100TB) fast (ssd) disk but that is just a vague recollection of another post somewhere

… and yes, people are being awesomely generous !!

[Jassu]

iirc WS need large (50-100TB) fast (ssd) disk

This can be arranged easily in a form of very well performing SAN as the storage backend.

Basically I would like to chat with someone who knows more about the Work Server tech, so we could figure out how feasible it would be to contribute to the project this way.

[ComputerUser]

Hi,

there is another thread where a researcher is listing these requirements:

Hi all! If you are interested in contributing hardware or server side power to our effort, please reach out to me at sukrit.singh@wustl.edu.
Generally the minimum requirements we have for setting up a work server are:
Debian Linux machines with 8GiB RAM, 100TiB SSD storage, and 1GiB/s network.

If you think you could help out please send me an email! Thanks so much for your support and help!

Maybe you can write him for details.

Cheers,

ComputerUser