TABLE OF CONTENTS
Updated: 18 Nov 2024
NVIDIA H100 GPUs On-Demand
Today, we are going to share our monthly product update for June. At Hyperstack, we're constantly pushing the boundaries of AI innovation and are excited to share our recent breakthroughs with you. This monthly update brings a new range of exciting features and improvements designed to enhance your cloud computing experience to new heights.
From supercharged networking capabilities to streamlined deployment processes, our Tech team has been working diligently to improve every aspect of our platform. But that's not all - we're also giving you a sneak peek into the future with our upcoming features that promise to revolutionise how you work with containers, GPUs and more.
Latest Hyperstack Features
Check out our recently released features to improve your experience on Hyperstack:
Bandwidth Boost for Network Performance
We've taken a giant leap forward in networking capabilities with our new High-Speed Ethernet Network. Bandwidth speeds range from 37 to 350 gigabits per second (Gbps), depending on thread count and MTU size, with a peak bandwidth of 400 Gbps. This significant upgrade is perfect for data-intensive workloads and high-performance computing tasks. Contracted users can request this feature through our technical support team for enhanced performance.
Want to see the numbers in action? Check out our detailed benchmarks and case study in our latest blog posts below:
- SR-IOV Series Part 2: Technical Details of RDMA, RoCE and Performance Benchmarks
- SR-IOV Series Part 3: Meta Case Study- Balancing RDMA and Ethernet for HPC
Deployment Made Easy
Our new one-click deployment feature comes with optimised default settings, making it easier than ever to get started.
Live-Chat Support
Need 24/7 assistance? Our new Hyperbot live-chat support is here to help. Connect instantly with our support team right from your console. We're committed to providing you with quick and efficient support to keep your projects moving forward.
Hyperstack R13 and R14 Release Highlights
Check out our platform with new features including:
- Multiple API Keys: We've implemented support for multiple API keys, giving you improved security and greater flexibility in managing access to your Hyperstack resources.
- YAML Support for Cloud-Init: You now have more power over your instance configurations thanks to our new YAML support for Cloud-Init.
- Improved Billing Transparency: In the 'Billing' section under the 'Resources' tab, you can now view the date and time of resource creation, providing you with more detailed insights into your usage.
- Streamlined Resource Cost Information: We've added separate tabs for 'Current Resources' and 'Historical Resources' in the 'Billing' section under 'Resources'. This makes it simpler to distinguish between current and past resource usage costs.
- Enhanced Resource Management: Pages displaying lists of resources now feature pagination and filtering options. This includes 'My Virtual Machines', 'Environments', 'Firewalls', 'Volumes', and 'Key Pairs', making it easier to navigate and manage your resources.
For a complete list of new features and improvements, be sure to check our release notes in our Infrahub documentation.
Upcoming Hyperstack Features
We're not stopping here! Here's a sneak peek at what's coming on our platform in July:
- Managed Kubernetes: We're launching the beta testing phase for our Managed Kubernetes service. Learn more about this service here.
- Terraform Provider and Go SDK: Beta testing for our Terraform provider and Go SDK will begin soon. Learn more about this service here.
What’s Next?
We will soon launch game-changing features and resources that will revolutionise your AI and data workflows including:
On-Demand NVIDIA H100 SXM and NVIDIA DGX GH200
We're working on making NVIDIA H100 SXM and NVIDIA DGX GH200 GPUs available on-demand by request. These high-end solutions will accelerate your most demanding AI and HPC workloads. To help you make informed decisions, we have comprehensive performance benchmarks for NVIDIA H100 SXM and NVIDIA DGX GH200 that allow you to choose the perfect GPU for your specific needs.
High-Performance Object Storage
Our upcoming high-performance object storage solution promises to deliver super-fast performance with speeds up to 250 MB/s per TB. This means you'll be able to access and process your data at unprecedented speeds without breaking the bank.
Container-as-a-Service
This user-friendly platform will allow you to deploy and run AI containers such as Jupyter Notebook and Stable Diffusion without requiring in-depth Docker knowledge. Our goal is to make AI more accessible to all so you can focus on innovation rather than infrastructure management.
We're constantly enhancing our platform to meet your needs and drive innovation. Your feedback is highly invaluable to us, so please don't hesitate to share your thoughts and suggestions at support@hyperstack.cloud. Thank you for being a part of the Hyperstack community. We can't wait to see what you'll build next!
Get started with Hyperstack
Watch the video below to explore our platform!
Subscribe to Hyperstack!
Enter your email to get updates to your inbox every week
Get Started
Ready to build the next big thing in AI?