spark-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Vadim Semenov <va...@datadoghq.com>
Subject Re: Writing a DataFrame is taking too long and huge space
Date Fri, 09 Mar 2018 15:02:52 GMT
You can use `.checkpoint` for that

`df.sort(…).coalesce(1).write...` — `coalesce` will make `sort` to have
only one partition, so sorting will take a lot of time

`df.sort(…).repartition(1).write...` — `repartition` will add an explicit
stage, but sorting will be lost, since it's a repartition

```
sc.setCheckpointDir("/tmp/test")
val checkpointedDf = df.sort(…).checkpoint(eager=true) // will save all
partitions
checkpointedDf.coalesce(1).write.csv(…) // will load checkpointed
partitions in one task, concatenate them, and will write them out as a
single file
```

On Fri, Mar 9, 2018 at 9:47 AM, Deepak Sharma <deepakmca05@gmail.com> wrote:

> I would suggest repartioning it to reasonable partitions  may ne 500 and
> save it to some intermediate working directory .
> Finally read all the files from this working dir and then coalesce as 1
> and save to final location.
>
> Thanks
> Deepak
>
> On Fri, Mar 9, 2018, 20:12 Vadim Semenov <vadim@datadoghq.com> wrote:
>
>> because `coalesce` gets propagated further up in the DAG in the last
>> stage, so your last stage only has one task.
>>
>> You need to break your DAG so your expensive operations would be in a
>> previous stage before the stage with `.coalesce(1)`
>>
>> On Fri, Mar 9, 2018 at 5:23 AM, Md. Rezaul Karim <
>> rezaul.karim@insight-centre.org> wrote:
>>
>>> Dear All,
>>>
>>> I have a tiny CSV file, which is around 250MB. There are only 30 columns
>>> in the DataFrame. Now I'm trying to save the pre-processed DataFrame as an
>>> another CSV file on disk for later usage.
>>>
>>> However, I'm getting pissed off as writing the resultant DataFrame is
>>> taking too long, which is about 4 to 5 hours. Nevertheless, the size of the
>>> file written on the disk is about 58GB!
>>>
>>> Here's the sample code that I tried:
>>>
>>> # Using repartition()
>>> myDF.repartition(1).write.format("com.databricks.spark.
>>> csv").save("data/file.csv")
>>>
>>> # Using coalesce()
>>> myDF. coalesce(1).write.format("com.databricks.spark.csv").save("
>>> data/file.csv")
>>>
>>>
>>> Any better suggestion?
>>>
>>>
>>>
>>> ----
>>> Md. Rezaul Karim, BSc, MSc
>>> Research Scientist, Fraunhofer FIT, Germany
>>>
>>> Ph.D. Researcher, Information Systems, RWTH Aachen University, Germany
>>>
>>> eMail: rezaul.karim@fit.fraunhofer.de
>>> <andrea.bernards@fit.fraunhofer.de>
>>> Tel: +49 241 80-21527 <+49%20241%208021527>
>>>
>>
>>
>>
>> --
>> Sent from my iPhone
>>
>


-- 
Sent from my iPhone

Mime
View raw message