FYI I'm from Packet. Agreed reliability is important (I'll let others comment on our own record) but I'd say there is significant workload via spot market, batch processing, etc that doesn't need uptime and is interested in cost.
Also, I've seen most users want 99.9+% uptime. 2% downtime is 14 hours in a short month. People would probably get angry about that...
Any chance you guys are going to offer something that is on the lower end of the spectrum for ARMv8 in terms of pricing? For eg. Scaleway has ARMv7 servers for 3 EUR/month (https://www.scaleway.com/pricing/)
I'll keep an eye on the Packet website. I did try out the ARMv8 server, a bit disappointed that the hardware did not support backwards compatibility with ARMv7 though :(
If Lustre is the answer, you're probably asking the wrong question. Unless that question involves short life-span, massively parallel swap-outs from one weapons simulation to another.
I write this not as storage religion (of which there's far too much), but to warn away those who haven't experienced the many kinds of data (and stomach lining) loss that come with being a Lustre admin.
Have a look at Open vStorage (http://www.openvstorage.com). Some highlights: open-source, core is battle proven for more than 7 years, performance, scales across datacenters , unlimited snapshots, ...
So $0.05/hr is pretty cheap, but agreed free trial is better. We made a little page with $25 on it if you'd like to try it out! https://www.packet.net/promo/hacker-news/
The clients of ours that leverage the by-the-hour model generally use the same devops tools (Terraform, Docker Machine, Ansible, etc) against our API to provision and orchestrate bare metal that they would against AWS or DO to do the same with VM's.
The use case drivers for those who choose bare metal often seem to be price/performance ratios, network, bring your own hypervisor or not use one, etc.
For anyone who actually would like to know how modern machine boots, in a way which is practical and not just theoretical, this article is probably the best there is.
Definitely recommended for someone coming from the legacy BIOS boot background, expecting UEFI to be the same. (Hint: It isn't)
Basically while legacy booting has too many black boxes and black magic (from an end-user perspective) to be practical to work with, UEFI is actually inspectable and debuggable. If something fails, you can figure out why.
Best of all: Creating live USBs/CDs/whatever no longer requires special tools. Just copy the everything, including the EFI folder, and its automatically bootable.
I've met Rishi @ Tiingo a few times and the fact of the matter is he is super passionate about liberating financial data and tools. I think that his pricing model and generosity are truly born from his genuine interest and love of the product. Seems strange, but its true! Hey anything to break up the 2.5k/mo bloomberg stranglehold....
