Comment by devy | Hacker News Reader

Comment by devy | Hacker News Reader

Parent Full thread

I agree your sentiments here. The goal for Alphabet is to make Cloud platform business profitable not shutting it down. From this article it looks like it hasn't been so since 2014 after GOOG gave it full attention. And to be profitable, Alphabet execs thought it has to be within top 2 to be profitable, I am unsure if that's true, given the cloud business is still a slow and painful transition process for Fortune 1000 businesses(some more advanced than the others).

Been using AWS for 8+ years since their first inception of S3, then 6 years of GCP. This year I had first hand experience with Azure. I just happen to think GCP has the most cutting edge cloud native technologies and services (GKE is much better than EKS or AKS, IMO; BigQuery is awesome). But AWS has a first mover's advantage and MSFT has the dominant enterprise customer base so pushing additional add-on sales is much easier. Those are GCP are up against.

Perhaps the market is big enough to accommodate top 3 players to be profitable. And GCP needs to do a better job in support their product & services not discontinuing them abruptly.

Google has cool things in GCP, but I struggle to imagine them being _popular_, cool, things.

If your product or project is big enough to require the scaling and capacity that GCP offers then you’re also probably big enough to be risk-averse about predicating your product’s viability on a platform you don’t control. The way I see AWS and Azure’s success: they replaced server colocation, dedicated server rentals and shared hosting for application hosting because they eliminated the costs and burden of systems administration and hardware maintenance. If you’re a company that did move to AWS/Azure - provided you stuck to their commodity offerings (App Service, EC2, RDS, Azure SQL, etc) you can move back to your own hardware just fine. For other things like S3/Azure Storage which are hard to move in-house, at least it’s easy to separate that from the rest of your application.

But what does GCP have that is cool - but without associated costs of what is essentially vendor lock-in? Without a solid contract/guarantee from Google that some binary I wrote that uses a GCP-exclusive service _will_ be supported for a minimum of 2-5 years it’s hard for me to justify risking it.

Google is not dumb and realizes this, which is why they're pushing so many tools that prevent vendor lock-in. K8s entire pitch is you can seamlessly go from a cloud offering to your locally managed K8s in a datacenter. Our (small) company has done it, it works very smoothly.

Their 'serverless' batch/stream processing works this way as well, with Apache Beam you can deploy effortlessly to GCP for large compute jobs (i've spun up hundreds of nodes in minutes), and also run it against on prem spark clusters with zero code changes.

So Google realizes being late to the game means they have to offer something more, and what they're offering is a lack of vendor lock in.

I feel it would be enough for Google to become a customer to GCP for it to become profitable.

I imagine the Search/Ads SRE team would tell the GCP team that they can have their infrastructure when they pry it out of their cold dead hands.

Reply on news.ycombinator.com