Re: Flink 1.6 Yarn Session behavior

2019-02-18 Thread Jins George
Thank you Gary. That was helpful. Thanks, Jins George On 2/17/19 10:03 AM, Gary Yao wrote: Hi Jins George, Every TM brings additional overhead, e.g., more heartbeat messages. However, a cluster with 28 TMs would not be considered big as there are users that are running Flink applications on tho

Re: Flink 1.6 Yarn Session behavior

2019-02-17 Thread Gary Yao
Hi Jins George, Every TM brings additional overhead, e.g., more heartbeat messages. However, a cluster with 28 TMs would not be considered big as there are users that are running Flink applications on thousands of cores [1][2]. Best, Gary [1] https://flink.apache.org/flink-architecture.html#run-

Re: Flink 1.6 Yarn Session behavior

2019-02-14 Thread Jins George
Thanks Gary. Understood the behavior. I am leaning towards running 7 TM on each machine(8 core), I have 4 nodes, that will end up 28 taskmanagers and 1 job manager. I was wondering if this can bring additional burden on jobmanager? Is it recommended? Thanks, Jins George On 2/14/19 8:49 AM, Ga

Re: Flink 1.6 Yarn Session behavior

2019-02-14 Thread Gary Yao
Hi Jins George, This has been asked before [1]. The bottom line is that you currently cannot pre-allocate TMs and distribute your tasks evenly. You might be able to achieve a better distribution across hosts by configuring fewer slots in your TMs. Best, Gary [1] http://apache-flink-user-mailing-

Re: Flink 1.6 Yarn Session behavior

2019-02-12 Thread Tzu-Li (Gordon) Tai
Hi, I'm forwarding this question to Gary (CC'ed), who most likely would have an answer for your question here. Cheers, Gordon On Wed, Feb 13, 2019 at 8:33 AM Jins George wrote: > Hello community, > > I am trying to upgrade a Flink Yarn session cluster running BEAM > pipelines from version 1

Flink 1.6 Yarn Session behavior

2019-02-12 Thread Jins George
Hello community, I am trying to upgrade a Flink Yarn session cluster running BEAM pipelines from version 1.2.0 to 1.6.3. Here is my session start command: yarn-session.sh -d -n 4 -jm 1024 -tm 3072 -s 7 Because of the dynamic resource allocation, no taskmanager gets created initially. Now