Hi Naresh,

May i know what kind of exceptions were thrown by the bolts ? How did you
fix the problem which you faced...your inputs might help me in correlating
with my issue.

--Padma Ch.


On Wed, Mar 26, 2014 at 8:16 PM, Naresh <[email protected]> wrote:

> Hi Padma
> It is hard to guess what problem you are facing.
> I can share a problem that we faced earlier on that was slowing the
> processing. You might have a similar issue.
> Our bolts were throwing exceptions which was causing restart of our
> workers. This slowed down the processing. After we fixed all our bolts the
> topology is working at great speed bounded only by external resources.
> Hope that helps.
> Regards
> Naresh
>
> On Mar 26, 2014, at 5:58 AM, padma priya chitturi <[email protected]>
> wrote:
>
> Hi,
>
> When using storm to process 1GB data file (has approx 200 million tuples),
> the execution proceeded at first while at a later point, processing has
> been stuck and the workers have restarted processing the file from the
> beginning. Does the issue lie with IO ?  As storm can process million of
> tuples per second, 200 million tuples wouldn't be an issue. Can someone
> suggest me on this...
>
>
> Thanks,
> Padma Ch
>
>
> On Wed, Mar 26, 2014 at 4:13 PM, Swara Desai <[email protected]> wrote:
>
>> Hi,
>> I am evaluating storm for some processing that might take upto one hour.
>> This implies that the bolt would be processing one tuple for an hour.
>> Can storm handle cases which have long processing time? Other than
>> configuring things like timeout, have any of you noticed any issues with
>> heavy processing?
>>
>> Thanks in advance
>>
>
>

Reply via email to