Hacker News Reader

Comment by kolanos | Hacker News Reader

I'm not a fan of these pricing models for VPSs. If you're fine with an oversubscribed OpenVZ VPS with minimal RAM/HDD then you might as well be on a shared virtual host. Even with an oversubscribed OpenVZ infrastructure, I can't imagine Atlantic.net makes a profit at this price point.I shudder to think where they're cutting corners. I can only assume the strategy is to get them (the customers) in the door at $0.99/mo. and hope they'll upgrade when they quickly outgrow their bare minimum VPS specs. But if I'm in that position, am I really going to stay with a OpenVZ provider instead of moving on to a KVM/Xen one where I know I'm getting what I paid for?

View on HN

OpenVZ's only going to get more painful once all the major distros migrate to systemd. The systemd developers basically assume that you're going to upgrade the kernel in lockstep with the rest of the distro and that's just not feasible with OpenVZ.

Are these OpenVZ? That's a shame.

Really, all I've ever wanted is an IaaS provider that gives me VMs in the shape of Heroku's instance partitions: a little bit of guaranteed CPU/memory/IOPS per dollar, with cheap/free low-jitter bandwidth thrown in, made specifically for allowing network-server applications that scale horizontally.

My 64MB-of-memory IO-bound Erlang chat server doesn't need a 4GB instance with eight cores. But it has to take that instance, because that's what you pay for in order to get a good uplink.

Its KVM, not OpenVZ.

Now that you mention it, it seems odd that companies don't let you tweak specs individually. RAM, CPU, disk space, network speed, etc. are all independent variables. Why not have some pricing formula and then give us a slider for each dimension and let us choose what we want?

I suppose that's because in they end, they want to allocate all the RAM, CPU, and network link available on each physical host. If a host has a 1Gb uplink, 32 cores and 256GB of RAM, and someone takes 1Gb bandwidth, 1 cores and 1GB of RAM, all the other cores can't be sold any more. It's easier for them to just split each host server in n VMs than to do some fancy stuff to try to maximize allocation.

That being said, some providers let you tweak everything independently. I know Gandi does [1], there are probably others.

(I don't use them, no idea what their cloud service is worth, I'm just showing the sliders)

[1] https://www.gandi.net/hosting/iaas/buy

I wonder if this is what contributes to the large price difference - for example, a 1Gb/1core Gandi instance is about 2x the price of the corresponding Linode plan.

Very nice, just the sort of thing I had in mind. I can see how the resource allocation problem would get really hairy, though. I wonder how Gandi deals with it.

Gandi and Cloud Sigma both do this. You get sliders where you adjust exactly what you want to allocate: Memory, disk, CPU speed, cores, network.

They do.. but the trouble with allocating and utilizing all available bare metal ram/cpu/bandwidth is clearly built into their price at the lower end

1 core, 512MB, 20GB ssd = $19.84/month or $13.89/mo prepay compare to DO at $5/month

4 core, 8GB, 80GB ssd = $124/month or $86.88/mo prepay compare to DO at $80/month

Anywho.. multiple reliable hosting options always good :)

I'd guess part of it is that it's easier to use homogenous hosts, and chop them up evenly.

Reply on news.ycombinator.com