This is for tenant-isolated k8s clusters using NVIDIA's DSX reference stack, relying on NICo / BlueField to set up isolated networking, and wipes/re-attests each host between tenants. Mostly useful if you're chasing NVIDIA Cloud Partner status or need NVIDIA support across the whole stack.
eel 2 hours ago [-]
> Mostly useful if you're chasing NVIDIA Cloud Partner status or need NVIDIA support across the whole stack.
What if I don't care about the cloud partner status or support, but I still have a lot of Nvidia hardware I need to hand out to run isolated workloads? Seems like deploying straight k8s is the way to go?
itintheory 49 minutes ago [-]
I'm not super familiar with this tool, but I am a k8s administrator. I think it depends on how isolated you need those workloads to be. This seems like a good fit if you want cluster level isolation. There are other projects that implement the Cluster API (CAPI) though, which provide similar functionality, eg declaratively managing tenant clusters.
websap 6 hours ago [-]
Its great to see them open source this stack!
Lots of negativity from HN posters who usually manage 3 Kubernetes clusters in a single region via a managed service on AWS or GCP.
5 hours ago [-]
debarshri 5 hours ago [-]
I got lot of hate too when I tried building a api server compatible engine in rust [1]
I am seeing a lot of openstack around since Broadcom enraged the vast majority of their vmware customers
Datagenerator 8 hours ago [-]
Just deployed OpenStack by using the Kolla Ansible installer. Works great coupled to Ceph for storage. Thanks to Broadcom we are moving in the right direction. Proxmox as fallback plan but it looks like OpenStack is solid.
9 hours ago [-]
Yasuraka 9 hours ago [-]
This seems to mix several (should-be) distinct concerns like CAPI, Kamaji and gpu-operator.
The only things I'd take from this is the kind-based bootstrapper for the management cluster and SPIFFE/TPM integration.
asaiacai 7 hours ago [-]
yeah i wish more visibility would go towards CAPI, Kamaji.
tbh i'm not sure from reading this why you wouldn't just roll rke2/k3s if the concern is an environment outside of the well worn hyperscaler managed k8s
This is what happens when AI generates your PR – FAQ. Describe what the product does versus just a data dump.
chandlerh 11 hours ago [-]
I’m usually in agreement with this sentiment, but they do say what it is in the first paragraph.
“NVIDIA Kubernetes® Engine (NVKE) provides the management plane that platform teams install to create and operate Kubernetes clusters for their customers.”
hedora 11 hours ago [-]
Ok, so say I have a platform team to install k8s or k3s. Why do I need NVKE? Also, how do I resell my kubernetes to customers? Like is there stripe integration?
I am more confused by its description than by just reading the product name and guessing.
Foobar8568 9 hours ago [-]
Well one needs to read a bit more to understand what that meant...
NVKE is not a Kubernetes distribution and not a Cluster API flavour. It is a self-hosting factory for Kubernetes clusters: a throwaway seed cluster stands up a management cluster, and the management plane on it then provisions and lifecycle-manages many isolated workload clusters on demand.
adastra22 11 hours ago [-]
That doesn’t say anything. It doesn’t answer a single question I would have.
Renina 5 hours ago [-]
[flagged]
topspin 11 hours ago [-]
A mention of something Kubernetes related reached the front page! A few years ago, BAI, there was one or more k8s headlines gracing HN's front page at all times.
Kinda glad that ended. k8s is fine and all, but it was a bit much for a while there.
luipugs 11 hours ago [-]
Did you just make an acronym that means "before AI"?
topspin 11 hours ago [-]
I did, but I'm not the first. I thought maybe I was, so I checked: apparently it's been bouncing around informally among academics for a while now. I hadn't previously seen it though, so I can at least say I came up with it independently.
I come up with things from time to time. At least one US registered trademark is mine.
wannabe44 11 hours ago [-]
Future history books will be like: ancient legacy technology conglomerate Google was founded around 24BAI
qlte 10 hours ago [-]
I'd guess us throwing around the word "AI" constantly and endlessly debating p(doom) based on current tech will be considered pretty amusing in the future.
Like reading essays from the late 19th century that can't stop pontificating about living in a boundless technological singularity thanks to railroads and the telegraph. Current hackocalypse discourse already has shades of "the bomber always gets through" from the 20s.
"Sure, people in every era since the Renaissance have foolishly believed they were the ones living on the cusp of uniquely transformative and potentially cataclysmic technological change but ... This Time It's Different"
HeWhoLurksLate 5 hours ago [-]
Or something like 32BAGI
iririririr 10 hours ago [-]
pretty sure "before wwiii" will take precedence
luna_whale42_ 9 hours ago [-]
[flagged]
Rendered at 15:21:07 GMT+0000 (Coordinated Universal Time) with Vercel.
What if I don't care about the cloud partner status or support, but I still have a lot of Nvidia hardware I need to hand out to run isolated workloads? Seems like deploying straight k8s is the way to go?
Lots of negativity from HN posters who usually manage 3 Kubernetes clusters in a single region via a managed service on AWS or GCP.
[1] https://github.com/debarshibasak/superkube
The only things I'd take from this is the kind-based bootstrapper for the management cluster and SPIFFE/TPM integration.
tbh i'm not sure from reading this why you wouldn't just roll rke2/k3s if the concern is an environment outside of the well worn hyperscaler managed k8s
“NVIDIA Kubernetes® Engine (NVKE) provides the management plane that platform teams install to create and operate Kubernetes clusters for their customers.”
I am more confused by its description than by just reading the product name and guessing.
NVKE is not a Kubernetes distribution and not a Cluster API flavour. It is a self-hosting factory for Kubernetes clusters: a throwaway seed cluster stands up a management cluster, and the management plane on it then provisions and lifecycle-manages many isolated workload clusters on demand.
Kinda glad that ended. k8s is fine and all, but it was a bit much for a while there.
I come up with things from time to time. At least one US registered trademark is mine.
Like reading essays from the late 19th century that can't stop pontificating about living in a boundless technological singularity thanks to railroads and the telegraph. Current hackocalypse discourse already has shades of "the bomber always gets through" from the 20s.
"Sure, people in every era since the Renaissance have foolishly believed they were the ones living on the cusp of uniquely transformative and potentially cataclysmic technological change but ... This Time It's Different"