Hacker News Reader

Comment by chucky_z | Hacker News Reader

I simply don't believe the $250 - $20,000 month comparison. Could you please provide some real-world numbers/hardware/stack information?

I just did this migration (non-AWS -> AWS), and we saw a ~20% reduction in monthly prices with 1-year contract at AWS vs. physical hosting (admittedly we were getting terrible pricing with our physical hosts, but still). This will be reduced with the addition of t2.nano servers as we run enough t2.micro instances for little odds and ends that it will put a dent in the bill.

View on HN

Your parent is [rightfully] including bandwidth. If he was able to get 1gbps unmetered, then the math is in the right ballpark (that's a huge if though).

Still, I agree with the parent's general sentiment re price. The gap used to be much worse, but it's still anywhere from 2x-20x depending on what you're doing.

- EC2 prices are simply high and exclude things like bandwidth.

- The CPUs are usually 1-3 generations behind (they've gotten better with this).

- Virtualization adds overhead (again, this can be very significant if you're doing a lot of system calls)

- IO options are significantly worse

- Their network connectivity is, at best, average

You said it yourself, you're comparing a "terrible" deal to AWS and only saved 20%. Sounds to me like AWS is only 20% better than terrible ;)

This is correct. The biggest cost I saw was bandwidth, followed by cycles on their machines. MediaCrush was a "media hosting website" (see original comment) and as such, used a lot of bandwidth and did a lot of work transcoding things, and had a large storage requirement.

Were you comparing bandwidth/data transfer in your numbers for EC2, S3 or CloudFront?

I don't know, it was an anecdotal story from a while ago. Before we left AWS we were utilizing all three (iirc these numbers didn't factor in CloudFront though). Point I'm making is - do the research and don't just reach immediately for AWS.

Oh okay I was just curious. I completely agree. The AWS bill at my startup went up quite a bit after the free trial ended since there are many variables that are included with each service such as how many requests/mo, tiered bandwidth, etc. The AWS calculator is nice and so is the billing estimator, but it's still quite confusing.

I think in this case I could have interchanged "terrible" with "enterprise" and it would've had a more ""correct"" meaning (seriously I wish I could keep adding quotes to all of that...). I had to fight and argue with the previous host (Rackspace) to get them to lower our prices. It also took weeks to deploy a server. Now it takes 5-25 seconds with a prebuilt AMI that I've cooked together, deployed on exactly the hardware that I want, with the network configuration that I setup, and the security rules (bi-directional! Rackspace firewall only offers inbound at the level we had) that I want/need. On top of this, to get lower prices, I open a ticket and say "Hello, I'll be needing this server for 1 year, lower prices please." No phone calls involved. I'm not trying to diss Rackspace, they seem to be doing a lot of cool things... I simply had a permanantly sour taste left in my mouth. AWS is a big juicy BBQ sandwich in comparison.

On your points about AWS itself I'm not going to argue with as they certainly are valid, however in my specific setup (everything production is c4/r3 types, 1tb gp2 volumes as all volumes, and running network benchmarks to ensure everything is properly sized) I've actually had a pretty big performance gain over dedicated servers in a DMZ behind a hardware firewall.

With regards to the bandwidth, it seems a common issue is when you have very chatty services (Apache Kafka) deployed across multiple AZ's the bill gets real big real fast. Thus far, we're not even cracking 3-digits in bandwidth monthly. Maybe we're simply not at a scale to notice these problems yet.

Reply on news.ycombinator.com