ficant, we might be
>> encountering an IO bottleneck. In this case, we could try increasing the
>> parallelism of this vertex, or, as Penny suggested, we could try to enhance
>> the memory of tm.
>>
>>
>> --
>> Best！
>> Xuyang
>>
>>
&g
Hi Ashish,
Can you check a few things.
1. Is your source broker count also 20 for both topics?
2. You can try increasing the state operation memory and reduce the disk
I/O.
-
- Increase the number of CU resources in a single slot.
- Set optimization parameters:
- task
2023 02:38:11 -0800 (PST)
MIME-Version: 1.0
References: <1678530886.3557992.ez...@flink.apache.org>
In-Reply-To: <1678530886.3557992.ez...@flink.apache.org>
From: Penny Rastogi
Date: Sat, 11 Mar 2023 16:08:00 +0530
Message-ID:
Subject: Re: confirm subscribe to user@flink.apache.or
