Comment by bmir-alum-007 | Hacker News Reader

Comment by bmir-alum-007 | Hacker News Reader

Parent Full thread

Not really. Amazon as a whole is dial-a-yield depending on the rate of profit reinvested to grow existing/new capabilities. If Bezos wanted, he could halt expansion and return more profit in the short-term, but that would delay long-term revenue (and hence profit) generation plans (however long he's able to find other business expansion directions which avoid the perils of overexpansion and also avoid plateauing for too long).

Also, some companies like Netflix and Zynga, which are AWS'es main customers, are fine to pay AWS a premium to have most of their infrastructure better managed and on-demand. Whereas most sizable businesses deploy something like OpenStack for baseline load on real, bare metal gear they buy and use AWS for traditional uses: PoCs, disaster recovery and peaking capacity. Startups tend to use AWS at first because it's easy, popular and low-risk, but it's not a panacea because it tends to get very expensive, very quickly. (It is however, clearly, a great business in itself, which probably should be spun-off.)

Disclaimer: I was with a premier enterprise AWS external consultancy shop early on that had significant access to capabilities and expanded limitation perks beyond other shops. Interestingly, they've built limitation increase requests available to most everyone now, which were manual before and depended on how much clout a customer or vendor garnered.

> it's not a panacea because it tends to get very expensive, very quickly.

I've investigated colocs near some AWS locations and found AWS to be around 20% cheaper than the cheapest viable option. With AWS's occasional price drops on top of that and general flexibility it's not a difficult choice.

One thing I found was that most providers were trying to sell vertical reliability (e.g. hardware RAID1), with a matching price. Our redundancy is horizontal at the application level (e.g. Zookeeper, Cassandra) so this has no value for us.

The bare-metal stuff I manage typically end up at 1/3 to 1/2 of AWS costs. Yes, fully loaded costs including salaries and office costs for anyone touching the servers.

You can certainly find more expensive options than AWS if you look, but similarly paying less than for AWS is easy too unless you start relying on specific AWS features.

Though, as always, your mileage may vary. Those costs have an awful lot of variables.

Is that rented gear in someone else's racks or your own gear in racks you rent? Beyond a few boxes, the TCO of buying your own gear is far, far cheaper (look at the prices of Supermicro boxes, not luxury, overpriced enterprise boxes that aren't use-case appropriate).

Someone also has to know how to negotiate with DC's (because power circuits are often the biggest cost factor) and how to separately get good deals on fiber (which is usually far cheaper than the DC's charges because most customers don't know better). DC's often nickel and dime for simple tasks if you don't do it yourself (or have hands that can do it for you). (RackSpace used to charge something $75 every time a box need to be reimaged. DC's will often take forever and charge several arms and legs to rack & stack & cable in your cage, but you will still be at their mercy for freight & delivery and pulling outside fiber, electrical circuits, cooling & rack leveling.)

Overall though, it's still cheaper for web-ops shops to do bare metal in (Google, Yahoo, Apple, etc. in their own DC's and other large/medium shops with predictable baseline loads in someone else's DC's) otherwise everyone would be using AWS. And that's clearly not the case for anyone who's ever been in a real enterprise shop.

Look at Wikia or most any other enterprise shop for baseline loads: mostly bare metal, because it's far cheaper for people whom know what they're doing. Also, BackBlaze... pretty much no way to undercut with them apart from buying an HDD manufacturer and going multi-datacenter-scale.

You'll always be able to engineer a better system for a particular known purpose.

The magic of AWS is that the transactional nature of it gives you more flexibility. I was working with somebody with about a storage requirement that will be in the 800-900TB range. But to hit the price point they want, they'd be buying a 1PB capacity solution.

The catch is, what's next? If they exceed their growth projection (a strong possibility), they need to spend $$$ in capital funds to buy some large block of capacity. That procurement takes time. Implementation requires more planning.

With Amazon, the prices are high/competitive, but the marginal cost of a unit of capacity is the cost of that capacity!

On the other hand, this user also has a large transaction system with well known performance an scaling characteristics. In their case, their buying power and size makes it much cheaper (by 3x) to host in a datacenter with customer-owned equipment.

Rented gear in someone else's racks, and it's only that cheap due to Supermicro. I did calculate what the providers premium was, and it didn't seem too unreasonable given the services such as remote hands and taking care of replacements that were being provided.

Perhaps in your use-case it was cheaper / less risk to rent than to own, but for sizable concerns, it's a no brainer that owning is far, far cheaper.

(For smaller stuff/side projects: In the UK, ByteMark.co.uk is awesome, flexible (colo/rent) and cheap. In the US, Pair.com is great. Linode is also decent.)

(For bigger stuff, there are datacenters in the middle of nowhere where power is cheap. That's the often limiting factor for green-field projects... cost of renting circuits from DC's, not the rack real-estate.)

I am brazilian and I am assuming OP is too, Amazon prices in Brazil are cheaper than the competition, simply because the local oferrings are really expensive.

Reply on news.ycombinator.com