er"
Gesendet: Freitag, 17. Juli 2020 05:41:23
Betreff: Re: Performance test Flink vs Storm
>From this exercise , I understand that increasing JVM memory would directly
>support/increase throughout. Am i correct?
It depends. Smaller heap space means more frequent GCs, which occupie
- It sounds like the CPU processing time is not the bottleneck.
>>>>Thus increasing the parallelism will not give you better performance,
>>>> but
>>>>will on the other hand increase the overhead load on the task manager.
>>>>
>>>> Also pulled in Becket
cket Qin, who is the expert of Kafka connectors. Since
>>> you have observed lack of performance in reading from Kafka compared to
>>> Storm.
>>>
>>> Thank you~
>>>
>>> Xintong Song
>>>
>>>
>>> [1]
>>> https://ci.apach
ing from Kafka compared to
>> Storm.
>>
>> Thank you~
>>
>> Xintong Song
>>
>>
>> [1]
>> https://ci.apache.org/projects/flink/flink-docs-release-1.11/ops/memory/mem_tuning.html#heap-state-backend
>>
>> On Thu, Jul 16, 2020 at 10:35 AM Prasan
Sending to you all separately as you answered one of my earlier query.
>>
>> Thanks,
>> Prasanna.
>>
>>
>> -- Forwarded message -
>> From: Prasanna kumar
>> Date: Wed 15 Jul, 2020, 23:27
>> Subject: Performance test Flink vs Stor
Hi,
We are testing flink and storm for our streaming pipelines on various
features.
In terms of Latency,i see the flink comes up short on storm even if more
CPU is given to it. Will Explain in detail.
*Machine*. t2.large 4 core 16 gb. is used for Used for flink task manager
and storm supervisor
