-
# Multiple GPUs
Currently, even though multiple GPUs are assigned to the project we don't see any time improvements in the training.
Actually we see that the time gets _worse by ~1 second_ for each…
-
Title: Decentralized Computing for Partitioning and Solving Hugging Face Models
Abstract:
This research proposal presents a novel approach to partitioning and solving Hugging Face models in a dece…
-
From #39, recovery testing really only makes sense as a semi-subcategory of performance testing, and the distinctions between its different definitions aren't really meaningful. This should be made ex…
-
#### Bug description
In scenarios where a single server installation is recovered from an image of the server into a new machine with different IP, services do not work because they are configured …
-
Since the storage array is a RAID-6 rather than ZFS or BTRFS, there should be regular integrity checks to verify the integrity of the data. This can be done by creating an md5 one the file is uploaded…
nanch updated
11 years ago
-
### What happened + What you expected to happen
While adding the `restore` function for [fault tolerance](https://docs.ray.io/en/latest/tune/tutorials/tune-fault-tolerance.html) when using the `TuneB…
-
We have documentation articles under `docs/articles` folder, many of them include code snippets.
Instead of having code pasted directly into articles, need to reference the code samples using DocFx…
-
Hi guys im newbie and i do not know if it can be done or not but it worth to ask.
I installed partkeepr on online ubuntu server , but i need another version (same database and .. ) on my raspberry …
-
We’re running fairly long tasks via MBrace, hours per job. It’s working really well for getting all the needed dependencies spread onto workers.
We have had some difficulties around cancellation. By …
-
Manages adding new commits to a ledger.
- Consensus
- Initiate commits
- Notify listeners (one of which is a listener)