All geek questions in one place

Elastic parallelism and fault tolerance in distributed Julia

How Julia reveals fault tolerance - when a node falls (intentionally or not) and when communication between nodes decreases.

I saw a few mentions of such a function, but I could not find out exactly how to do this.

+4
source share
1 answer

In the pmap dockstrongs, you can see that this is already implemented there using the keyword arguments retry_.

pmap([::AbstractWorkerPool], f, c...; distributed=true, batch_size=1,
on_error=nothing, retry_n=0, retry_max_delay=DEFAULT_RETRY_MAX_DELAY,
retry_on=DEFAULT_RETRY_ON) -> collection

... pmap . , on_error, , . , , , inline .

retry_on, retry_n, retry_max_delay, retry_on, n max_delay . , , .

, - @parallel. Base.wrap_on_error Base.wrap_retry . , pmap https://github.com/JuliaLang/julia/blob/v0.5.0/base/pmap.jl.

, (, , ), , , . .

+3

Source: https://habr.com/ru/post/1671738/

More articles:


All Articles