4 ms·
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 an
by DaiPlusPlus 7y ago
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.
- runT1ME 7y agoGoogle 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.