Assuming the "spinnaker" it talks about is https://spinnaker.io/ how does it translate the move from virtualization to deployment?
Is it being implied that Telefònica Germany is moving to the cloud, and dismissing its vmware license (and hardware, and datacenter stuff) ? Was Telefonica running vSphere on the Managed VMWare offering on AWS?
Don't get me wrong, I loathe BroadCom as much as the next guy, but this article isn't very informative.
This makes sense for the "bidders" part of the article. They need to slowly transform things and slowly desert/transform systems.
And I don't think this involves moving anything other than support responsibilities.
This article is just about buying support from another provider.
VMware does _a lot_ more than just virtualization these days. This most likely has something to do with Tanzu Platform for Cloud Foundry or just Tanzu, which does provide CI/CD services.
Which they likely migrated to Spinnaker and vanilla Kubernetes.
There are effectively only three mobile providers in Germany. 1&1 is the fourth, but its network doesn't even cover the whole country. From those three networks, Telefonica is the smallest by quite a margin. So the sentence in the article is technically correct, but draws a nonsensical picture.
Could. But not in Germany.
Just look at the parallels in wired broadband, which are equally lacking behind some European peers.
That is only true if you count M2M sim cards. Looking at regular postpaid contracts, the picture is actually quite different:
Deutsche Telekom: 26.8m customers
Telefonica: 26.2m customers
Vodafone: 19.3m customers
I've seen some countries where they had three established GSM networks and then someone starts repurposinf their CDMA network for mobile handsets, but it had been built and operating for years for fixed wireless, phone lines for houses without wires.
However, a few years down the line, Tele2 ended up in a merger with T-Mobile (now Odido), and Ziggo/UPC/Vodafone went through mergers to end up as VodafoneZiggo.
You're talking like two years is long term...
For the current Broadcom leadership this doesn't have to be sustainable, just bring in enough of a profit to justify the whole thing. They do not want a slow and steady profit stream but a big squeeze and done. I bet they got what they hoped for.
but i think the 'eventually' is the reason this is working for them
[1] https://blogs.vmware.com/cloud-foundation/2024/11/11/vmware-...
[2] https://knowledge.broadcom.com/external/article/399823/vmwar...
I suspect the strategy is to extract maximum cash whilst they can.
Personally I don't have a huge amount of experience with either, so I'm possibly a bit biased, but I see k8s as either a hyperscale solution or someone wanting to be cool and trendy.
Same goes for serverless, other than niche use cases of say running python code in a 100% Azure ecosystem when you just have absolutely zero other choice.
It may not be k8s as we know it today, but many SMEs are most certainly using containers, via Docker.
Ultimately containers use less resources than a full VM and allow dependency management.
VMs is a reducing business.
I'm aware it's technically possible to have a large, wide fleet of containers...but I don't really see it going that way (at all).
My current thoughts of why is separation of knowledge - IT knows how to manage infra and virtualisation, and the vendor/teams know how their individual docker containers need to be set up/work.
I do agree full virtualisation is probably on the decline, but the current tech is definitely staying around in my size of business.
Oracle is mostly used on servers. /s
ojosilva•6mo ago