Yes. It has the same effect as `x=0`. It just give a safer value to `x`.

On Thursday, August 13, 2015 at 6:28:11 PM UTC+2, Marcio Sales wrote:
>
> x would still be listed.
>
> Em quinta-feira, 13 de agosto de 2015 10:32:01 UTC-3, Sisyphuss escreveu:
>>
>> Is `x=nothing` a good idea, since `nothing` has no meaning and could be 
>> tested `x==nothing`?
>>
>>
>> On Thursday, August 13, 2015 at 1:13:09 AM UTC+2, Tony Kelman wrote:
>>>
>>> Good question. I have not tried to do this, it's been about 6 years 
>>> since I last used a Windows cluster.
>>>
>>> Built-in, it looks like the only remote cluster manager available by 
>>> default in base Julia communicates over ssh - 
>>> http://docs.julialang.org/en/latest/manual/parallel-computing/#man-clustermanagers
>>>
>>> Is your cluster using Windows HPC Server? How would you usually create 
>>> parallel jobs and communicate between workers? Are you using Microsoft MPI? 
>>> There is an open pull request for adding support for Microsoft MPI to 
>>> MPI.jl (see https://github.com/JuliaParallel/MPI.jl/issues/51 and 
>>> https://github.com/JuliaParallel/MPI.jl/pull/52), but I don't think 
>>> it's actually been tested across multiple machines yet. Maybe support for 
>>> Windows HPC Server could be added to the ClusterManagers.jl package? Would 
>>> have to check the documentation both for ClusterManagers.jl 
>>> https://github.com/JuliaParallel/ClusterManagers.jl and Windows HPC 
>>> Server https://technet.microsoft.com/en-us/library/ff919397.aspx to see 
>>> what would be needed.
>>>
>>> I don't think there's an easy way to delete a variable binding. You can 
>>> call workspace() which really just moves the current content of your Main 
>>> module to Main.LastMain.
>>>
>>> On Wednesday, August 12, 2015 at 1:32:58 PM UTC-7, Marcio Sales wrote:
>>>>
>>>> Hello all. New user here with a couple' questions. 
>>>> Is it possible to set up Julia for distributed computing in Windows 
>>>> cluster? How? 
>>>> Is there a way to delete a variable once created? Not just free mem 
>>>> making x=0. 
>>>>
>>>> Thank you. 
>>>> M.
>>>
>>>

Reply via email to