info_tfgrid/collections/tfgrid4/slices/aislice.md

34 lines
1.1 KiB
Markdown
Raw Normal View History

2024-05-23 08:09:37 +00:00
# AI Slice
2024-04-25 10:14:09 +00:00
2024-05-23 08:09:37 +00:00
!!wiki.include page:'slice_intro.md'
2024-04-25 10:14:09 +00:00
2024-04-26 17:52:37 +00:00
An AI box is a unit of AI capacity (GPU or future TPU driven).
2024-04-25 10:14:09 +00:00
2024-04-26 17:52:37 +00:00
- The mininal GPU supported for now is a Nvidia 4090 or comparable
2024-04-25 10:14:09 +00:00
An AI box can be launched in our Zero-OS and can enable any possible AI workload.
## AI Hour (AH)
2024-05-23 08:09:37 +00:00
AI hour (AH) is like a kwatth unit for electricty: it represents a AISlice being used for 1h and billed as such.
2024-04-25 10:14:09 +00:00
INCA Hosts (our cloud providers) can price the AI Hour themselves in a chosen currency.
2024-05-23 08:09:37 +00:00
## AISlice Properties
2024-04-25 10:14:09 +00:00
2024-05-23 08:09:37 +00:00
The service provider (hoster) defines the following properties per AISlice:
2024-04-26 17:52:37 +00:00
- Type of GPU
- Price per hour
- Discounts if used for longer periods & if renter has large amount of TFT in wallet
- Max bandwidth
- Min bandwidth (min 1 mbit/sec)
- Cost per GB bandwidth if any
2024-05-23 08:09:37 +00:00
- Max aggregation size (how many of the AISlices can be combined in one node if relevant)
2024-04-26 17:52:37 +00:00
- Link to support site if any (find info about hoster and service capabilities)
- If linked to Hosting Pool (company giving support on the machines)
- Location & type of location
2024-04-25 10:14:09 +00:00
2024-05-23 08:09:37 +00:00
Each AISlice has unique ID and can be looked at through a portal (or found).
2024-04-25 10:14:09 +00:00