| ▲ | JackC 5 days ago |
| > I work for an org with close ties to arXiv, and just like us they are getting a lot more demand due to AI crawling Funny, I also work on academic sites (much smaller than arXiv) and we're looking at moving from AWS to bare metal for the same reason. The $90/TB AWS bandwidth exit tariff can be a budget killer if people write custom scripts to download all your stuff; better to slow down than 10x the monthly budget. (I never thought about it this way, but Amazon charges less to same-day deliver a 1TB SSD drive for you to keep than it does to download a TB from AWS.) |
|
| ▲ | Imustaskforhelp 5 days ago | parent | next [-] |
| I don't understand, why don't you use cloudflare? Don't they have an unlimited egress policy with R1? Its way more predictable in my opinion that you only pay per month a fixed amount to your storage, it can also help the fact that its on the edge so users would get it way faster than lets say going to bare metal (unless you are provisioning a multi server approach and I think you might be using kubernetes there and it might be a mess to handle I guess?) |
| |
| ▲ | sitkack 4 days ago | parent | next [-] | | Regardless, if you are delivering PDFs, you should be using a CDN. If crawling is a problem, 1 it is pretty easy to rate limit crawlers, 2 point them at a requestor pays bucket and 3, offer a torrent with anti leech. | |
| ▲ | mcmcmc 5 days ago | parent | prev [-] | | Could have something to do with Cloudflare’s abhorrent sales practices. | | |
| ▲ | keepamovin 5 days ago | parent [-] | | Can you tell me more? I think my business needs some abhorrent sales practices. That's how it's done, right? | | |
|
|
|
| ▲ | ryao 4 days ago | parent | prev [-] |
| The two are not comparable. The 1TB of transit at Amazon can be subdivided over many recipients, while the solid state drive is empty and only can be sent to one. That said, I agree that transit costs are too high. |
| |
| ▲ | fc417fc802 4 days ago | parent [-] | | So order multiple drives, transfer the data to them, and drop them in the mail to the client. That should always be the higher bandwidth option, but in a sane world it would also be less cost effective given the differences in amount of energy and sorts of infrastructure involved. The reason to switch away from fiber should be sustained aggregate throughput, not transfer cost. | | |
| ▲ | ryao 4 days ago | parent [-] | | The other guy was also comparing them based on transfer cost. Given that 1TB can be divided across billions of locations, shipping physical drives is not a feasible alternative to transit at Amazon in general. | | |
| ▲ | fc417fc802 4 days ago | parent [-] | | I'm not trying to claim that it's generally equivalent or a viable alternative or whatever to fiber. That would be a ridiculous claim to make. The original example cited people writing custom scripts to download all your stuff blowing your budget. A reasonable equivalent to that is shipping the interested party a storage device. More generally, despite the two things being different their comparison can nonetheless be informative. In this case we can consider the up front cost of the supporting infrastructure in addition to the energy required to use that infrastructure in a given instance. The result appears to illustrate just how absurd the current pricing model is. Bandwidth limits notwithstanding, there is no way that the OPEX of the postal service should be lower than the OPEX of a fiber network. It just doesn't make sense. | | |
| ▲ | ryao 4 days ago | parent [-] | | That is true. I was imagining the AWS egress costs at my own work where things are going to so many places with latency requirements that the idea of sending hard drives is simply not feasible, even with infinite money and pretending the hard drives had the messages prewritten on them from the factory. Delivery would never be fast enough. Infinite money is not feasible either, but it shows just how this is not feasible in general in more than just the cost dimension. |
|
|
|
|
