Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Make installation path for Apple silicon users less complicated #127

Closed
victorlin opened this issue Aug 10, 2022 · 2 comments
Closed

Make installation path for Apple silicon users less complicated #127

victorlin opened this issue Aug 10, 2022 · 2 comments
Assignees
Labels
enhancement New feature or request

Comments

@victorlin
Copy link
Member

victorlin commented Aug 10, 2022

Context

Currently,

  1. We recommend "Native" runtime for Apple silicon users because emulated Docker is slow.
  2. Conda installation on Apple silicon has many quirks, and requires users that already have Conda installed to read through a lengthy FAQ page.

As people start using newer Macs, we will see more users being caught in this complicated installation path.

Possible solutions

  1. Wait until a full Nextstrain installation on arm64 Conda is possible. This means all of the following which currently don't support arm64/noarch must add support for that:

    1. Augur dependencies
    2. Auspice
    3. Other packages we include in the current installation, mostly to support running the ncov workflow.

    This could be achieved if bioconda adds support for compiling recipes on arm64, which is something @corneliusroemer brought up in Slack, but timeline is pretty much out of our hands here.

  2. Provide a new "Conda" runtime which gets set up and updated independently from any existing installation. This was @tsibley's idea mentioned during our installation discussion meeting. This would use the Intel installer for all Macs, in line with our current recommendation. Update: this is happening over at Managed Conda runtime cli#218.

  3. Provide "native" Docker support which will hopefully make Docker runtime speeds comparable to current Native runtime speeds on Apple silicon. This would require finishing off Build multi-arch image with amd64 + arm64 for M1 Macs docker-base#35 and doing some benchmarking.

@tsibley
Copy link
Member

tsibley commented Aug 23, 2022

2 and 3 are my preferred approaches here. I assume 1 may happen on its own (or become more feasible at least) with enough time.

@victorlin
Copy link
Member Author

Both solutions 2 and 3 have been completed, and the installation path is better as of #157.

@github-project-automation github-project-automation bot moved this from Prioritized to Done in Nextstrain planning (archived) Jun 20, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
No open projects
Development

No branches or pull requests

2 participants