Always good to see the "critics said x, but supporters said y" habit as one indicator of impartial reporting.

I don't get why people are in denial that Azure is doing well. If you want to know why Google isn't doing -great- in the cloud, it's because people are afraid they will kill the project. It's because they don't have a "hook" like AWS and Azure have. AWS has a massive user base already, good reason to adopt it. Azure hooks into so many other Microsoft services and developer tools, good reason to adopt it. GCS has...what? Sure, some cool things, but we're talking about compute, compute where where the money is. Even if you adopt GCS for some edge or AI projects, that's not a lot of spend.

Spot on, and it is a sentiment I come across more than once every year. The number of startups hosted on AWS is huge, MS Azure is currently runner-up and Google a distant third.

And while this may be splitting hairs, I think it matters what they're actually selling. If they're moving on-prem customers to SaaS, that doesn't seem like a particularly useful comparison, since we already know MS sells a tonne of proprietary software to enterprise.

I think the ambiguity that people are so inclined to over engineer that they can't possibly imagine something so simple as a vanilla VM with a SQL SaaS hooked to it as "the cloud" because it's not complex enough.

you can argue that running active directory in a cloud vm is still in the cloud - yes, but that's not the same kind of growth because that entire old windows market is static at best.

The best Kubernetes hosting?

As far as I’m aware, azure and office get reported separately. Azure is “intelligent cloud” and office (along with some others) is “more personal computing”

Google is still at 3% market share which puts it at the same level as IBM Soft Layer and Rackspace. [0] Given the technologies Google has developed for cloud-native/cloud-scale computing (Kubernetes, MapReduce, TensorFlow, etc.) not to mention outstanding apps like Gmail, this result has to be a disappointment to their investors.

[0] https://www.skyhighnetworks.com/cloud-security-blog/microsof...

Google had to invest massively to catch up to AWS purely technically, and it's only now that people are starting to consider it to be close. Then consider that companies will tend to pick a cloud provider and not revisit that decision for 5 or 10 years.

This has to be a very long-term play where the results won't show up for many, many years. Greene's tenure should be judged in terms of market perception (e.g. if we were making the decision now instead of 10 years ago, would Google be viable?) as opposed to a large absolute increase in market share (so check back 10 years from now).

1. AWS is the clear market leader, tons of services, good support options, but you'll need pretty expert-level ops people to make sure everything is working together securely. You may have to build more plumbing than you think. 2. Azure is great for companies that are already MS shops and it makes lots of MS tech (.Net, SQL Server) super easy to use in the cloud. 3. If I were starting a startup, I would go with GCP. IMO their services are easier to use with a smaller ops team, and they have lots of services that scale more easily as you grow (e.g. Cloud Spanner). But not as good enterprise-level support.

I see lots of excitement for GCP in the startup scene, and as these startups become more mature I think the "enterprise-level"-ness of GCP will improve.

I’m not entirely convinced that the F500 would adopt GCP if they had the support and hand-holding structure that they have historically demanded. Myself, I’d rather find a solution that doesn’t need such a team to support me in the first place. Sure, there’s all sorts of issues with partners’ service offerings and the high cost of transit between multiple cloud providers but perhaps you can save the costs of support and put it toward network instead (won’t solve the transit costs for your partners though).

It is being pushed by both outside consultants and employees who become the experts within the enterprise. GCP's ease of use is a disadvantage in the big picture.

What examples come to mind?

Enterprises tend to buy full IT offerings including apps and services, not just IaaS. Having a big partner network, which VMware does, is a competitive advantage. Amazon and Azure likewise have large partner ecosystems.

Disclaimer: I work for VMware.

A smart executive (ha!) would announce that next year’s bonus pool would be funded by savings on Oracle licences then sit back and let nature take its course...

https://news.ycombinator.com/item?id=16871931

Read the last description: "amateur and unprofessional." The same thing happens for Kubernetes Engine. Does Google internally run their Cadillac infra like this? Of course not. The conclusion is either they don't care about their external cloud or that team can't ship basic features.

Yes, on rare occasion there's an unplanned failure (host error) and the VM has to be restarted elsewhere, but this is the only option for maintenance at AWS.

Mandatory disclaimer: I work at GCP.

https://news.ycombinator.com/threads?id=flamingcow

This is pretty strong customer feedback. What happened with this feedback, anything? The bigger picture here about GCP, preceded by Google's long reputation, is that customer feedback means nothing. Googlers are probably still building what they feel like building, and that is why AWS continues to trot ahead despite being less polished. They ship things that work and they know what their customers want.

I've been a GCP customer for ~18 months and as recent as a year ago it still felt like very early days for sales and support. Well, actually it still feels like early days for support. The products, however, have been surprisingly good for the most part and seem more mature and well-engineered than their market share would suggest, especially in comparison to Azure.

So, I'm of the opinion that if they can figure out how to sell, they'll do quite well.

So what exactly is it that Google is doing now that will lead to long-term profitability in a business where revenue is closely tied to scale? Can Google really prevail over Amazon (first mover) and MS Azure (great enterprise chops + Office) based on capital investment alone? Amazon and Azure are not standing still. Both also have an order of magnitude advantage in market share. Finally GCS is not strong in the big growth markets in Asia.

For my money Google's best bet is to stay in the game and hope something alters the market in a way that weakens their competitors. That assumes they can double down on the investment to remain viable and recognize the opportunity before others do.

Second, Google is clearly betting that its scale will indeed give it a benefit -- the scale of its data centers which include all of Search and everything else (e.g. raw costs for computing and storage being cheaper than competitors), the scale of its investments in machine learning which it makes available on GCP, the scale of its investments in private networking and security, etc. Part of Google's pitch is that it's providing Google-class security and networking which AWS and Azure simply can't match.

They would need to grow their marketshare ten-fold to be in that position, and they would need to defeat Alibaba on its home turf. It’s not going to happen. GCP could possibly drive IBM and Oracle out of the cloud market, but even that’s unlikely.

There is an extra dose of disappointment because on technical merits, GCP's basic VM substrate -- Compute Engine and Cloud Storage -- is far superior to AWS even five years ago, but AWS kept their ear to the street, anticipated every managed service and ancillary feature that their customers could have wanted, and executed relentlessly. Google has not mounted even half as vigorous and urgent a response as needed under the circumstances.

There is nothing at the outset that would prevent someone from choosing GCP, many might even be predisposed given the reputation of Google's own infrastructure, but when more and more of the tasks that you might want to offload to cloud are still not available or not reliably working on GCP, and you're stuck with operating VM's, you might seriously consider the competitor, and once that decision is made, you are not coming back.

It's an execution, and perhaps cultural, problem with Google, not one of technology or market.

Cloud computing seems to me like an industry at high-risk of being commotitized after the initial gold rush -- similar to smartphones. If that's the case, I'm not sure it makes sense to prioritize the future over today.

For instance, our app uses Postgres 10's native partitioning features, it's REALLY strange to me that Cloud SQL is still on PG 9.6. They also don't communicate effectively on when / if they'll add it which makes planning difficult. AWS on the other hand is already testing PG 11 and has an open schedule in place to their customers can plan ahead.

The leaner than AWS comment was when he made the decision in the past and the close call comment was more recent.

The question isn’t whether you or I are happy with Goohle’s performance but how management and investors see it.