Scaleway Blog

> a TWh, which is about six million years' of power i'm not a physics guy but wanted to check this - a Watt is a per second measure, and a Terawatt is 1 trillion watts, so 1 TWh is 50 billion seconds of 20 Watts, which is 1585 years of power for a single brain, not 6 million. i'm sure i got this wrong as i'm not a physics guy but where did i go wrong here? a more neutral article (that doesn't have a clear "AI is harming our planet" agenda) estimates closer to 1404MWh to train GPT3: https://blog.scaleway.com/doing-ai-without-breaking-the-bank-yours-or-… i dont know either way but i'd like as best an estimate as possible since this seems an important number.

Would have loved to use Hetzner, sadly they rejected a picture of my ID card/passport (with some details blacked out) a few years ago when i tried registering and verifying my identity, i don't think there was any process for appealing to have someone else look at it again, thus my account ended up being blocked. Of course, this is just 1 data point and i'm sure that the experience of other people is more positive in general! That said, the prices actually look pretty competetive and are indeed cheaper than Scaleway's offerings. However, Scaleway also recently came out with more cost effective instances, called Stardust (though the specifications are somewhat limited): https://blog.scaleway.com/a-star-is-born-as-scaleway-launches-stardust… Personally, i use Time4VPS for my VPSes, which is a somewhat small daughter company of a larger Lithuanian telecom (Interneto vizija). Affiliate link, if anyone is interested in having a look: https://www.time4vps.com/?affid=5294 I've been using them for a few years, the uptimes and the performance both are generally pretty good, can even self-host an e-mail server and the few IPv4 addresses i've had from them haven't been blacklisted anywhere. However, they don't directly compete with SaaS or PaaS like AWS's or Scaleway's managed offerings.

The reason I have submitted the migration document instead of official announcement is because there isn't any as of writing. I received a mail few hours back informing[1] that the ARM64/C2 instances would be retired by Dec 1 and technical support will end on Jul 1. The reason for the retirement is vague, the email mentions 'several stability issues'. I've personally been running one of my projects on ARM64[2] for nearly a year and haven't had any major issues except reboot require power cycling. The announcement sounds fishy, it's not the first time Scaleway has retired its low tier instances[3], I suspect it do more with their Business than actual stability. I've recommended Scaleway's ARM instances several times, even here on HN; But now I'm reconsidering my options as I've been using ARM for my compute needs for a while now to align with my carbon goals. Update: Official forum discussion[4], looks like others feel the same way. [1] Someone else have posted the email content on Reddit - https://www.reddit.com/r/selfhosted/comments/g11uuj/scaleway_retiring_… [2]https://needgap.com [3]https://blog.scaleway.com/2018/introducing-scaleway-nextgen-nvme-cloud… [4]https://community.scaleway.com/t/c2-arm64-instances-end-of-life/8677

They are definitely not the first https://www.ikoula.com/en/dedicated-server/raspberry-pi-4-micro-server Scaleway was the first to offer ARM based servers (non dedicated) https://blog.scaleway.com/scaleway-become-the-first-arm-based-ubuntu-p… (2015)

Yep! They launched the C1 servers in 2013 and in 2014 they were providing Docker support for them [1]. Of course, those were simple and slow chips and these are 80cores Ampere Altra, but of course, it's not the first dedicated ARM service in Europe. Note: IIRC Scaleway had their C1 servers booting from the network, not local disk, but still... 1: https://blog.scaleway.com/docker-on-c1/

That's for the exact same reason (obvious identification and easy scanning) that I proposed to prefix all Scaleway's access keys with `SCW`. See: https://blog.scaleway.com/strengthening-scaleway-token-security-policy…

Chia coin is using Proof of Space. It is growing really fast and not a really good thing. Even cloud providers are reacting to it. See https://blog.scaleway.com/scaleway-and-chia/

According to the blog post "every service, feature and scenario currently provided via its DEV1-S instances will also be available on Stardust Instances" [0]. The DEV1-S is an AMD Epyc and the technical specs match up as well. So, that seems most likely. [0] https://blog.scaleway.com/a-star-is-born-as-scaleway-launches-stardust…

They are part of the working group https://blog.scaleway.com/2020/gaia-x-catalog-demonstrator-graph/

C2 introduced in April 2017 it appears: https://blog.scaleway.com/2017/scaleway-disruptive-armv8-cloud-servers… The C1 was introduced in 2013. And now Scaleway's Arm experiment is over?

>> the physical servers hosting them are indeed randomly affected by several stability issues this may cast a shadow on ARM64 servers credibility. they used to design their first generation of arm servers in-house. it looked like a blade server with slottable cpu(or was it the whole system on a micro-board?). check the video from 2014(at the end) for some glimple on a physical machine https://blog.scaleway.com/2014/online-labs-public-preview/?fbclid=IwAR…

> None of my ISPs (home, office and mobile carrier) currently offer IPv6, which I am personally frustrated about, but I totally understand why they don't. Can you get around this by using a tunnel/VPN to a provider [0] which does allow IPv6? [0] https://blog.scaleway.com/2016/introducing-native-ipv6-connectivity-on…

If you want to see more than 2 pictures (one of which is of a small drainage channel), there's a detailed blog post about the construction: https://blog.scaleway.com/2016/c14-story-part-1-meet-our-nuclear-fallo…

The C2M is Intel Atom based (Avoton). https://blog.scaleway.com/2016/c2-insanely-affordable-x64-servers/

Traffic's a luxury tax (along with RAM) that cloud providers assume that big companies can afford to pay if they're getting that much traffic. Outside of the Cloud Providers Traffic is dirt Cheap, Hetzner includes 30TB traffic inclusive in their dedicated server i7 Quad-Core Skylake 64GB DDR4 RAM, 2x250 GB SATA 6 Gb/s SSD for 39 euro /month: https://www.hetzner.de/us/hosting/produkte_rootserver/ex41sssd If you don't want to be shaped after you exceed 30TB, Hetzner charges €1.17 per additional TB, so 700TB would come to €783.90 total. Whereas ScaleWay include unlimited traffic in their bare metal servers starting from 12 euro /month: https://blog.scaleway.com/2016/03/08/c2-insanely-affordable-x64-server…

I know nothing about Scaleway, but found these via the link above: https://blog.scaleway.com/tag/docker/ https://www.scaleway.com/imagehub/docker/ what is the problem with kernel? very curious as I would need Docker myself if I was going to use them

But what about bandwidth? That's still shared among all the users, right? And given that they are even advertising 'Seedboxes' and VPN servers [0] hosted in the same network, it makes me doubt the quality of network even more. [0] https://blog.scaleway.com/2015/09/02/we-are-slashing-the-c1-price-by-7…

> What sort of ARM cpu's are they? 200Mhz Palm Pilot, or modern 1.4ghz A9 ones? 4-cores ARMv7 CPU with 2GB of RAM and a 1Gbit/s network card source: https://blog.scaleway.com/2015/04/02/from-online-labs-to-scaleway/

Online Labs[0] turned into Scaleway. There's a blog post about it at [1] -- [0] http://labs.online.net/ [1] https://blog.scaleway.com/2015/04/02/from-online-labs-to-scaleway/

The timeline of events in Scaleway's blog post is very dubious. If the SSD was stolen over one year ago why do they only acknowledge it now? In March 2021 [1] they were writing about how great their security was: > We are proud of our data centers and their security. We consider that we have implemented the best solutions to protect your most valuable asset: your data. We are well aware of the huge responsibility this represents. There can be no compromises when it comes to your data. Why were customers whose data was leaked only informed in June 2021? The delay of over 1 year is a huge GDPR issue. The timeline of the incident from public sources: 21 May 2021: Micode tweets a screenshot of the directory listing [2] 26 May 2021: First video on the subject [3] 6 June 2021: Scaleway customer is notified their data was leaked [4] 21 July 2021: Second video on the subject [5] 24 July 2021: Third video on the subject [6] 24 July 2021: Scaleway releases a French blog post stating "Over a year ago, an SSD was stolen" [7] Storing the data unencrypted is bad, but IMHO Scaleway's handling of the incident creates much bigger questions about their credibility. [1] https://blog.scaleway.com/how-we-protect-your-data/ [2] https://mobile.twitter.com/Micode/status/1395640486715662336 [3] https://www.youtube.com/watch?v=vt8PyQ2PGxI [4] https://www.lowendtalk.com/discussion/comment/3258386/#Comment_3258386 [5] https://www.youtube.com/watch?v=aOBVZUL1iBA [6]…

The C2 bare metal offer is based on the Atom C2000 series, which has the AVR54 errata. They are based on in-house custom hardware [1] and are no longer manufactured. The switching fabric used in this offering only supports stateless firewalls, creating a huge issue of feature parity between offerings. The switch fabric is part of the hardware design, so it is impossible to replace it with something that would support dynamic firewalls. The ARM64 offering is based on ThunderX servers from Gigabyte / Cavium. These servers were plagued by kernel panics from the first kernels provided by Cavium and firmware bugs that had never been fully resolved before the servers went into production. They were also plagued by consumer-grade 1TB SSDs that were constantly crapping the bed and losing customer data (unlike the other 1.92TB enterprise SSDs installed, which were robust). Offer C1 is also based on custom hardware [2] and faces the same situation as offer C2. If you are an existing C1 customer, I would take a backup and consider alternatives. I don't find it very surprising that they announced the withdrawal of these offers, you can check their GitHub repositories to see years of neglect to these products: operating system images rarely updated [3], and major usability issues have never been solved [4]. (Throwaway account to protect sources) [1] https://blog.scaleway.com/2016/c2-insanely-affordable-x64-servers/ [2]…

They're using Atom-based (old 2008 pre-Silvermont Atom architecture) Avoton "microserver chips" that Intel launched in 2013: https://blog.scaleway.com/2016/03/08/c2-insanely-affordable-x64-server… http://www.v3.co.uk/v3-uk/news/2292691/intel-launches-avoton-as-atom-c… If you want cheap VPS with good performance and reliability you may want to take a look at these as well: http://vpsdime.com/ https://www.ovh.com/us/vps/vps-ssd.xml https://www.prometeus.net/site/vps.php https://www.prometeus.net/site/kvm-vps.php I don't think either are using Atom. VPSDime is just overselling, but I haven't heard anyone complain about been bottlenecked, so they probably do it in a smart way (taking resources from those who don't use them fully).