Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Google Cloud vs. AWS Onboarding Comparison

kevinslin.com

Post reply on HN
Print view View original

131–140 of 394 posts

Re: Google Cloud vs. AWS Onboarding Comparison

#131

Post by bitbuilder »

musha68k wrote: post #36

From personal experience in multiple startups I can confirm that with regards to customer service and sales processes AWS is light years ahead of GCP. That said I have heard even better stories about Azure actually - apparently it is yet in another league of its own in terms of perks and actual service game.

The Azure service game is indeed incredible. Every experience I've had with them has been excellent.

A client once requested we file a support ticket with Azure to help deal with a performance issue my team was working through. It wasn't really all that urgent, but the client requested we use the highest urgency level anyway. So I filed our ticket.

In less than a minute I felt like my phone was being blown up by every engineer at Microsoft. And the messages they left made it clear that the fate of humanity hinged on resolving our issue within the next five minutes.

On top of that, the depth and intelligence of the support was downright humbling to this fellow engineer.

(At the end of the day, it turns out we'd screwed up and left debug logging on.)

Re: Google Cloud vs. AWS Onboarding Comparison

#132

Post by unethical_ban »

Johnny555 wrote: post #54
doomslice wrote: post #11

Earlier quoted context omitted.

I was able to get 130k free GCP credits (over 3 years) just by filling out the startup forms... as a complete nobody. You don't even need an affiliate to sponsor you.

If you're planning on building your business on the cloud, the availability of free credits seems like the worst way to evaluate a cloud provider. Generous "free" credits might even be a warning that they have to pay people to use the platform.

This ignores the other big reasons to take losses early, when one is a new(er) player in a market with a dominant competitor.

It's not completely wrong, but incomplete.

Re: Google Cloud vs. AWS Onboarding Comparison

#133

Post by throwaway894345 »

ethbr0 wrote: post #107
byteofbits wrote: post #102

Earlier quoted context omitted.

We’re currently migrating to Spanner for a variety of reasons - but the mandatory downtime on their Postgres CloudSQL offering will be the part I miss the least. It’s insane that even with all of their HA and failover turned on they take the whole cluster down for as long as they like every few months!

One thing that customer obsession at an entire-organization-depth level does is encourage broad customer use awareness. To an engineer, things are things, because of how they architect and build them. To an engineer who understands a customer, things are things and all the things people actually use them for. Big difference. It also makes "Well, that customer is using it wrong" less of an exceptable engineering dodge…

Surely product makes these decisions, not engineers, right? I agree that customer empathy is important, but I don't think we can conclude that the engineering team (rather than the product team) is the source of the deficiency?

Re: Google Cloud vs. AWS Onboarding Comparison

#134

Post by dmlittle »

colde wrote: post #113
dmlittle wrote: post #97

Earlier quoted context omitted.

While you do pay for AWS support, I must say that in my experience AWS support is pretty top notch. I don't particularly like the (somewhat) recent changes where the priority of your ticket is based on your support plan but I'm guessing it's because everyone always chose "critical" when making small support ticket.

In my experience with AWS support, it's a major difference on whether you are asking EC 2 questions or some of the lesser used service questions. For Media services, the supporter will almost always need to coordinate with an internal team, which there is no visibility over, and then it becomes a game of telephone to make the supporter relay the information in a way the internal team understands. I've had the same th…

I guess that makes sense. The quality of their support is probably directly correlated to the level of internal tooling to help diagnose issues. For more popular/older services that tooling is probably better.

Re: Google Cloud vs. AWS Onboarding Comparison

#135

Post by cbushko »

llbeansandrice wrote: post #99
cbushko wrote: post #70

Earlier quoted context omitted.

Not the original poster but we migrated to GCP because: 1) AWS was extremely expensive 2) Our GCP bill is about 1/3 of what AWS was 3) The Kubernetes offering is top notch 4) Google giving us credits and offering us consulting were the triggers that started us talking.

Seems like k8s is a big seller for a lot of companies. We don't use it currently on our team and the larger company is whole-sale in on AWS I'm not sure they'd ever be able to make the switch. Always like to see why people make these big changes though, thanks!

I spent over a year doing a full migration from AWS EC2 + ECS instances to GCP + docker + kubernetes. It was a huge task that has paid off very well.

  1) Costs per customer are lower because you can fit more containers per VM due to kubernetes doing the scheduling for you. Customers also include developer environments.
  2) The number of deploys is way up because there is a simple and established pattern that everyone follows.
  3) The speed of creating new services has increased because of the established patterns with containers, kubernetes resources, and deploys. Thinks days vs weeks to get something running.
  4) The number of Ops issues are lower because kubernetes handles so many things for you. For example, if a deploy is incorrect for some reason, the old service is sitting there running. No outage = no escalation = everyone sleeps at night.
Even if I was a tiny startup, I would still recommend using Kubernetes. The patterns, tooling and insight that Kubernetes gives you will save you TIME. The time saved is worth more than the tiny cost of a 3 node Kubernetes cluster. That is time you can use to develop your product and sell it vs time spent ftp'ing binaries to your Digital Ocean instance. :)

Re: Google Cloud vs. AWS Onboarding Comparison

#136

Post by stevencorona »

eyal_c wrote: post #59
stevencorona wrote: post #30

I used to be a huge fan of GCP and bet on it to power my startup, and have come to greatly regret it. Recently, I needed to increase a CPU-limit quota from a small number (like 16 vCPUs to 64 vCPUs) - nothing crazy. In the past, the quota increase system was more or less automated and would only take a few minutes to process. This time, however, GCP denied my quota increase and forced me to schedule a call with a sal…

I just started using Google Cloud SQL – the allure of a managed Postgres service was strong. Can you share some of your experiences with it?

Sure.

- No way to upgrade major postgres version without full export and import into new cluster.

- Incredible delay between postgres versions. IIRC, it took nearly 2 years for them to add postgres 11 after it was released.

- HA is basically useless. Costs double, still has 4-5 minute window of downtime as it fails over, doesn't avoid maintenance window downtime (both primary/standby have same maintenance window) and you can't use it as a read replica. Honestly, feels like a borderline scam since I'd imagine a new instance could be spun up in the same amount of time a failover takes (but I haven't tested)

- With default settings, we experience overly aggressive OOM-killer related crashes on a ~monthly basis during periods of high utilization. On a 32GB instance, OOM killer seems to kick in around 27-28GB and it's incredibly annoying.

- Markup over raw instances is almost 100%, with no sustained use discount outside of a yearly commit.

It's just a lot of money to pay for a crashy, outdated version of Postgres.

Re: Google Cloud vs. AWS Onboarding Comparison

#137

Post by cbushko »

stevencorona wrote: post #120
cbushko wrote: post #62

Earlier quoted context omitted.

CloudSQL was slow for us until we do the following: 1) Increase the disk size to 100GB as this increases the IOPs 2) Switch to using private IP addresses. Huge speed increase 3) get rid of cloudsql-proxy. Another huge speed increase These 3 things have kept our database instances very small and costs low.

Yeah, have hugely over provisioned disks for IOPS. Am still using public ip + cloudsql-proxy because the alternative didn't exist when I first deployed, but I'll try switching.

I went through this during last summer. The nice thing is that you can switch to private ip and cloudsql-proxy will still work. At least you can isolate your changes.

Re: Google Cloud vs. AWS Onboarding Comparison

#138

Post by cbushko »

ransom1538 wrote: post #117
cbushko wrote: post #62

Earlier quoted context omitted.

CloudSQL was slow for us until we do the following: 1) Increase the disk size to 100GB as this increases the IOPs 2) Switch to using private IP addresses. Huge speed increase 3) get rid of cloudsql-proxy. Another huge speed increase These 3 things have kept our database instances very small and costs low.

3) get rid of cloudsql-proxy. Another huge speed increase ^ Do not use cloudsql-proxy ever. GCP docs are wrong. DO NOT proxy all your db requests through a single VM.

Ours cloudsql-proxies were on running on GKE so they were not "that bad".

Switching to private ip definitely had the largest impact by far on performance.

Re: Google Cloud vs. AWS Onboarding Comparison

#139

Post by jjoonathan »

coder543 wrote: post #128
jjoonathan wrote: post #21

Earlier quoted context omitted.

Exactly. Google: gives me $300 of credit no questions asked and tells me what things will cost upfront. AWS: makes me apply to a program to get credits and if I want to price out anything it's almost a full-blown research effort where I have to dig through documents and cross reference tables between different services and I still probably miss something important. Free tier is opaque enough to frustrate even "use it…

I don't really agree with that comment. I think all of the megaclouds have undesirably complex pricing structures, but they also all provide pricing calculators. AWS has one here: https://calculator.aws/ If I punch in the services I want to use, I can quickly see how much they will cost, and it's easy. I recommend clicking "Advanced" inside services on the calculator if you want to be sure you're not going to run int…

I don't really agree with this comment, because we've gone through year after year of new AWS cost analysis and pricing tools that always seem to miss the mark in a major way that's only obvious in hindsight and always seem to be consistently worse than GCP and Azure. They never seem to fix the flaws in old tools, they just tack on new tools with new blindspots. Hence my comment about cross-referencing. If you're willing to put in a lot of effort across tools, you can get a complete picture, but it really does take a lot of effort and foresight into the exact structure that your solution is going to take (which involves cross-referencing documentation). Amazon doesn't make any effort to quote you a price once they have enough information, they always make you work for it. If a price transparency tool is gated behind enough effort, is it really a price transparency tool?

That said, I'll grant you that AWS is not the only megacloud leveraging opaque cost structures. They just leverage them more.

AWS support is genuinely a cut above, though.

Re: Google Cloud vs. AWS Onboarding Comparison

#140

Post by coder543 »

jjoonathan wrote: post #139
coder543 wrote: post #128

Earlier quoted context omitted.

I don't really agree with that comment. I think all of the megaclouds have undesirably complex pricing structures, but they also all provide pricing calculators. AWS has one here: https://calculator.aws/ If I punch in the services I want to use, I can quickly see how much they will cost, and it's easy. I recommend clicking "Advanced" inside services on the calculator if you want to be sure you're not going to run int…

I don't really agree with this comment, because we've gone through year after year of new AWS cost analysis and pricing tools that always seem to miss the mark in a major way that's only obvious in hindsight and always seem to be consistently worse than GCP and Azure. They never seem to fix the flaws in old tools, they just tack on new tools with new blindspots. Hence my comment about cross-referencing. If you're wil…

> pricing tools that always seem to miss the mark in a major way that's only obvious in hindsight and seems to be consistently worse than other clouds.

Can you provide an example of a service that the AWS calculator doesn't compute the correct price for?

I honestly can't remember ever being surprised by what things on AWS cost, and I don't think I'm that shockingly good at detecting hidden costs.

Post reply on HN