banyan-julia
banyan-julia copied to clipboard
banyan-team
A suite of familiar Julia APIs for bigger datasets with less cloud and lower costs.
This PR significantly reduces time for session startup and also latency of retrieving results of computation (by not always waiting for the "evaluation end" message). Depends on #147
This PR makes starting sessions lazy (so that cloud computing resources are only created when actually needed) and improves the API in other ways. Separate PRs will be needed to...
This makes the sampling feature of Banyan support multiple languages, formats, Banyan clusters, and more. It's critical to supporting Banyan Python and sampling use-cases. It also parallelizes downloading samples from...
Error: ``` Preparing session with ID 2022-06-14-20005104694a67276b791fecd9c7c898766798 Time: 0:00:51┌ Error: Cloning into 'banyan-julia'... │ HEAD is now at 94d5c30 Add debug statement │ Already up to date. │ Activating project...
We should get this in the next release.
Right now, we have a `configure_scheduling` that allows us to enforce parallelism such that replications is disallowed for values with memory usage > 0. This means that only things like...
The EC2 SSH key pair should not be required when creating a new cluster. And when a user creates a new account, a default cluster should be created without any...
Some logs: ``` Sending EVALUATION_END for 2021-10-19-1345239a639451e0afd719059d54319a7d78c2 Finished sending EVALUATION_END message in 1 chunks In CopyFrom with loc_name=Value and loc_params=Dict("value" => Colon()) In CopyFrom with loc_name=Value and loc_params=Dict("value" => :sepal_length)...
Metadata
Owner
Metadata
A suite of familiar Julia APIs for bigger datasets with less cloud and lower costs.