Would love to have any suggestion or comments about our implementation.
Is there anyone who has such experience?
Thanks.
Best Regards!
-
Luke Han
On Tue, Oct 27, 2015 at 10:33 AM, 周千昊 wrote:
> I have replace default java serialization with Kyro.
> It indeed reduce the sh
I have replace default java serialization with Kyro.
It indeed reduce the shuffle size and the performance has been improved,
however the shuffle speed remains unchanged.
I am quite newbie to Spark, does anyone have idea about towards which
direction I should go to find the root cause?
周千昊 于2015年1
We have not tried that yet, however both implementations on MR and spark
are tested on the same amount of partition and same cluster
250635...@qq.com <250635...@qq.com>于2015年10月23日周五 下午5:21写道：
> Hi,
>
> Not an expert on this kind of implementation. But referring to the
> performance result,
>
> i
