Show HN: TensorDock Core GPU Cloud – GPU servers from $0.29/hr Hello HN! I’m Jonathan from TensorDock. After 7 months in beta, we’re finally launching Core Cloud, our platform to deploy GPU virtual machines in as little as 45 seconds! https://ift.tt/k6zM4vU Why? Training machine learning workloads at large clouds can be extremely expensive. This left us wondering, “how did cloud ever become more expensive than on-prem?” I’ve seen too many ML startups buy their own hardware. Cheaper dedicated servers with NVIDIA GPUs are not too hard to find, but they lack the functionality and scalability of the big clouds. We thought to ourselves, what if we built a platform that combines the functionality of the large clouds but made it priced somewhere between a dedicated server and the large clouds? That’s exactly what we’ve done. Built to make engineers more productive. We have 3 machine learning images so you can start training ML models in 2 minutes, not 2 hours. We provide a REST API, so you can integrate directly your code with ours. And, there’s a community CLI you can use to manage your servers directly via command line We provide a feature set only large clouds supersede. We have storage-only billing when the VM is stopped (for only $0.073/GB/month) so that you aren't paying for compute when you don't need it. We also provide the ability to edit virtual machines after they’re created to downsize costs. If you provision a NVIDIA A6000 and find out you’re only using 50% of it, stop the VM, modify it to a NVIDIA A5000, and you’ll be billed the lower hourly rate without needing to recreate your server and migrate data over! Our infrastructure is built on 3x-replicated NVMe-based network storage, 10 Gbps networking, and 3 locations (New York, Chicago, Las Vegas) with more coming soon! - CPU-only servers from $0.027/hour - NVIDIA Quadro RTX 4000s from $0.29/hour - NVIDIA Tesla V100s from $0.52/hour - and 8 other GPU types that let you truly right-size workloads so that you’re never paying for more than you actually need We're starting off with $1 in free credits! Yes, we sound cheap… but $1 is all you need to get started with us! That’s more than 3 hours of compute time on our cheapest configuration! Use code HACKERNEWS_1 on the billing page to redeem this free credit :) TensorDock: https://ift.tt/vHXdUGg Product page: https://ift.tt/k6zM4vU API: https://ift.tt/vX0byxh Community CLI: https://ift.tt/wm7069r Deploy a GPU: https://ift.tt/TmpVvLb I'm here to answer your questions, so post them below! Or, email me directly at jonathan@tensordock.com :) https://ift.tt/k6zM4vU July 29, 2022 at 02:28AM
Show HN: Launch VM workloads securely and instantaneously, without VMs Hello HN! We've been working on a new hypervisor https://kwarantine.xyz that can run strongly isolated containers. This is still a WIP, but we wanted to give the community an idea about our approach, its benefits, and various use cases it unlocks. Today, VMs are used to host containers, and make up for the lack of strong security as well as kernel isolation in containers. This work adds this missing security piece in containers. We plan on launching a free private beta soon. Meanwhile, we'd deeply appreciate any feedback, and happy to answer any questions here or on our slack channel. Thanks! April 29, 2021 at 07:50AM
Comments