[Bioc-devel] ‘SOCKcluster’ cluster error with bplapply() under merida2

[Martin Morgan]

Also FWIW changing things like (there are a number of these changes)

     sum(unlist(lapply(data, length)))/1e+06

to
     sum(lengths(data)) / 1e6

actually has measurable effect on the speed of your code.

Martin

On 04/18/2018 10:47 AM, Carmen M. Livi wrote:
> Dear Martin Morgan,
> 
> ok, I will switch back to parallel then. Thank you very much for your 
> reply.
> 
> Kind regards,
> Carmen
> 
> On 18/04/2018 15:30, Martin Morgan wrote:
>> I'm sorry not to respond to your first email. This is a problem in 
>> BiocParallel that I am working on. In your particular case, where you 
>> are relying on a third-party package that uses parallel::mclapply, I 
>> think you should revert to using that.
>>
>> Martin
>>
>> On 04/18/2018 09:22 AM, Carmen M. Livi wrote:
>>> Hi all,
>>>
>>> I was using parallel::mclapply() but have been encouraged to substitute
>>> it with BiocParallel::bplapply().
>>>
>>> BiocParallel::bplapply(data,
>>>       BPPARAM = BiocParallel::MulticoreParam(workers =x ),
>>>       FUN  =...)....
>>>
>>> I never had problems with the code. Now the package got accepted and
>>> moved to the bioconductor-repository and suddenly I am getting errors in
>>> the vignette build when asking for 5 workers:
>>>
>>> "Error in serialize(data, node$con, xdr = FALSE) : error writing to
>>> connection"
>>>
>>> After reducing the number of requested workers to 3 my package passed
>>> under linux (malbec2 and malbec1) but still fails under OS X (merida2):
>>>
>>> "Error : failed to stop ‘SOCKcluster’ cluster: error writing to 
>>> connection"
>>> https://master.bioconductor.org/checkResults/3.7/bioc-LATEST/ChIC/merida2-buildsrc.html 
>>>
>>>
>>> Does anyone have an idea what is going on? Am I not supposed to ask for
>>> more then 1 worker in the vignette?
>>>
>>> What if I just switch back to parallel::mclapply?
>>>
>>> Thank you,
>>> Carmen
>>> //
>>>
>>>     [[alternative HTML version deleted]]
>>>
>>> _______________________________________________
>>> Bioc-devel at r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>>>
>>
>>
>> This email message may contain legally privileged and/or confidential 
>> information.  If you are not the intended recipient(s), or the 
>> employee or agent responsible for the delivery of this message to the 
>> intended recipient(s), you are hereby notified that any disclosure, 
>> copying, distribution, or use of this email message is prohibited.  If 
>> you have received this message in error, please notify the sender 
>> immediately by e-mail and delete this email message from your 
>> computer. Thank you.
> 


This email message may contain legally privileged and/or...{{dropped:2}}