Status and upcoming services

Current status

All systems are operational, no maintenance is planned.

Recent updates

Hibernation and persistent storage

 

Dynamic storage volumes went live early May 2021!

The new setup allows the user to have data on a high performance NVMe cluster and/or a HDD cluster. By default, all data is kept in 3 duplicates for redundancy and performance. This brings us the flexibility to deliver exciting new features such as:

  • Hibernation/restoration in a matter of seconds, regardless of size
  • Multiple storage volumes per instance
  • Moving instances to different hardware
  • Moving any storage volume to any server

To enable networked storage, we connected our entire infrastructure to a 50Gbit networking backbone. While having user instance data on a local drive simplified our initial infrastructure, the network storage cluster in early testing proves to be faster than the onboard NVMe drive in pretty much any metric. It even outperforms any cloud provider we compared with in terms of sequential bandwidth and delivers 30k-40k IOPS!

API

We have a fully functioning API in public beta. Try it out! You can create credentials and find a link to the docs on the account info tab.

New GPU instances

We are no longer expanding on V100 servers. While the V100 still holds up as an excellent GPU even in 2021, we are moving to the latest gen gear. We have launched our first RTX A6000 48GB servers early May!

Roadmap

Persistent storage volumes

We're working on adding additional features to the current dynamic volumes;

  • Creating and storing custom images
  • Cloning instances

Upcoming GPU instances

We are mainly expanding into 2 GPU instance types; the RTX A6000 (GA102, 48GB VRAM, 300W) and the A100 (80GB VRAM, SXM4/NVLink, 400W).

Our A100 80GB cloud instances are currently in validation while we keep adding RTX A6000 servers.

The A100 80GB instances are the fastest 8 GPU servers in existence right now with a mindblowing 640GB of VRAM at 2.0TB/s memory bandwidth per GPU!

Additionally, we're adding CPU instances intended for transferring data and running inference services at lower cost.

Let us know what you are looking for!

Our main focus is you! Tell us what makes your life easier and how we can save your precious time! ie.

  • What kind of image would be useful? 
  • What kind of feature can we add?
  • What kind of instance best fits your needs?
Arrow-up