Cost of serving billions of images per month • Hacker News Board

Re: Cost of serving billions of images per month

#31

Post by sergiotapia » Sun, May 05, 2019, 2:21 AM UTC

tjwds wrote: post #24

manigandham wrote: post #9

What's the business model here? We would never run with such expensive options for this functionality, especially if it was a free/donations powered service. This company raised $7.5M though: https://techcrunch.com/2018/02/15/unsplash-simple-token-seri...

They also have extensive partnerships with certain brands — their homepage currently serves ads disguised as content from a handful of companies. (I say this as a user of and contributor to Unsplash)

So ads? I thought it would be something a little more inspired.

Re: Cost of serving billions of images per month

#32

Post by nine_k » Sun, May 05, 2019, 2:26 AM UTC

tpetry wrote: post #18

manigandham wrote: post #14

Earlier quoted context omitted.

And some projects really are just simple. Scale makes no difference to their services, it's a basic OLTP + search with resizing image proxy. This doesn't have any SLAs nor is it critical software either. 24/7 oncall is completely unnecessary.

Don‘t underestimate the massive cpu power needed for scaling images at their request rate

Maybe some caching would help.

Re: Cost of serving billions of images per month

#33

Post by systemBuilder » Sun, May 05, 2019, 2:34 AM UTC

10(+) billion images served monthly for $100k = 1/1000th of a cent per image. This is not bad but I think Netflix is 2c a gigabyte; you are 10c (I'm assuming 100k per image).

Re: Cost of serving billions of images per month

#34

Post by reggieband » Sun, May 05, 2019, 2:34 AM UTC

I'm curious what their price-per-GB is from Fastly/Imgix ... the monthly total doesn't really tell me the full story.

At the last place I worked I was in charge of a team that ran our image service so I have a very solid understanding of the infrastructure around it and what resources we had in place to develop/monitor/support/etc. It wouldn't be hard to do a calculation and see if they are using money inefficiently.

I will say that most people don't really have any idea of the cost of simply pushing bits. The majority of what I did was video distribution which gets into eye-watering costs pretty quickly. Without knowing a bit more detail on the precise volumes they are storing/moving I don't think it is fair to criticize.

Re: Cost of serving billions of images per month

#35

Post by marknadal » Sun, May 05, 2019, 2:35 AM UTC

First time hearing about unsplash, thank you!!! This is awesome for us OSS devs.

Re: Cost of serving billions of images per month

#36

Post by kitotik » Sun, May 05, 2019, 2:38 AM UTC

Mizza wrote: post #23

quickthrower2 wrote: post #5

Earlier quoted context omitted.

Moreover what if Heroku goes out of business? Then they’re f’d. Hosting on k8s, using Terraform or whatever would give them a lot of automation and provider independence.

Can we please, please, please kill the myth that Terraform in any way helps in any way with provider independence?

I find it much easier to maintain terraform code that allows me to use multiple cloud providers than if I had to do the same without terraform.

You seem to have some insight to share here though, could you expound?

*edit typo

Re: Cost of serving billions of images per month

#37

Post by subpixel » Sun, May 05, 2019, 2:40 AM UTC

manigandham wrote: post #9

What's the business model here? We would never run with such expensive options for this functionality, especially if it was a free/donations powered service. This company raised $7.5M though: https://techcrunch.com/2018/02/15/unsplash-simple-token-seri...

They raised money from the crypto-coin outfit run by the former fab.com CEO. At Fab and at his previous company it seems the modus oprandi was to raise millions and then pocket as much as possible before the shit hit the fan.

Re: Cost of serving billions of images per month

#38

Post by nemothekid » Sun, May 05, 2019, 2:42 AM UTC

dna_polymerase wrote: post #15

> We continue to use Heroku as our main web platform. Despite its premium cost over AWS, Azure, and Google Cloud, Heroku’s built-in deployment and configuration tools allow our team to move faster, more confidently, and more reliably. This is just bullshit, really. I'd buy this excuse from a service like Snapchat/Facebook in their early years, but unsplash is nothing that changes dramatically all the time. Every sing…

>They could easily save 60% on their monthly bill. That would enable them to hire 1 or 2 engineers capable of maintaining a few dozen servers on AWS, DO or GCP.

So blow up their currently working infra to save 40k to reallocate that 40k to build a replacement? How does that make sense? This idea seems penny wise, pound foolish - run the numbers yourself.

Re: Cost of serving billions of images per month

#39

Post by tschellenbach » Sun, May 05, 2019, 2:51 AM UTC

I work with these guys since they are customers of Stream. Super solid team, really respect how they have such focus in their execution.

Re: Cost of serving billions of images per month

#40

Post by abhiyerra » Sun, May 05, 2019, 3:00 AM UTC

tpetry wrote: post #18

manigandham wrote: post #14

Earlier quoted context omitted.

And some projects really are just simple. Scale makes no difference to their services, it's a basic OLTP + search with resizing image proxy. This doesn't have any SLAs nor is it critical software either. 24/7 oncall is completely unnecessary.

Don‘t underestimate the massive cpu power needed for scaling images at their request rate

Wonder if they could use S3 bucket with a lambda to resize. I’m sure their images are a long tail so certain images likely have way more traffic than others. At a previous company where we were also doing a lot of imgix traffic we did something like this but with a go service but now I’d just offload to s3 and use lambda to avoid Devops..