Hi,

I've launched JobManager and TaskManager on DC/OS successfully.
Now I have two new issues:

1) All TaskManagers are scheduled on a single node.
- Is it intended to maximize data locality and minimize network communication cost?
- Is there an option in Flink to adjust the behavior of JobManager when it considers multiple resource offers from different Mesos agents?
- I want to schedule TaskManager processes on different GPU servers so that each TaskManger process can use its own GPU cards exclusively.  
- Below is a part of JobManager log that is occurring while JobManager is negotiating resources with the Mesos master:
2018-01-09 07:34:54,872 INFO  org.apache.flink.mesos.runtime.clusterframework.MesosJobManager  - JobManager akka.tcp://flink@dnn-g08-233:18026/user/jobmanager was granted leadership with leader session ID Some(00000000-0000-0000-0000-000000000000).
2018-01-09 07:34:55,889 INFO  org.apache.flink.mesos.scheduler.ConnectionMonitor            - Connecting to Mesos...
2018-01-09 07:34:55,962 INFO  org.apache.flink.mesos.runtime.clusterframework.MesosFlinkResourceManager  - Trying to associate with JobManager leader akka.tcp://flink@dnn-g08-233:18026/user/jobmanager
2018-01-09 07:34:55,977 INFO  org.apache.flink.mesos.runtime.clusterframework.MesosFlinkResourceManager  - Resource Manager associating with leading JobManager Actor[akka://flink/user/jobmanager#-1481183359] - leader session 00000000-0000-0000-0000-000000000000
2018-01-09 07:34:56,479 INFO  org.apache.flink.mesos.runtime.clusterframework.MesosFlinkResourceManager  - Scheduling Mesos task taskmanager-00001 with (10240.0 MB, 8.0 cpus).
2018-01-09 07:34:56,481 INFO  org.apache.flink.mesos.runtime.clusterframework.MesosFlinkResourceManager  - Scheduling Mesos task taskmanager-00002 with (10240.0 MB, 8.0 cpus).
2018-01-09 07:34:56,481 INFO  org.apache.flink.mesos.runtime.clusterframework.MesosFlinkResourceManager  - Scheduling Mesos task taskmanager-00003 with (10240.0 MB, 8.0 cpus).
2018-01-09 07:34:56,481 INFO  org.apache.flink.mesos.runtime.clusterframework.MesosFlinkResourceManager  - Scheduling Mesos task taskmanager-00004 with (10240.0 MB, 8.0 cpus).
2018-01-09 07:34:56,481 INFO  org.apache.flink.mesos.runtime.clusterframework.MesosFlinkResourceManager  - Scheduling Mesos task taskmanager-00005 with (10240.0 MB, 8.0 cpus).
2018-01-09 07:34:56,483 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            - Now gathering offers for at least 5 task(s).
2018-01-09 07:34:56,484 INFO  org.apache.flink.mesos.scheduler.ConnectionMonitor            - Connected to Mesos as framework ID 59b85b42-a4a2-4632-9578-9e480585ecdc-0004.
2018-01-09 07:34:56,690 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            - Received offer(s) of 606170.0 MB, 234.2 cpus:
2018-01-09 07:34:56,692 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            -   59b85b42-a4a2-4632-9578-9e480585ecdc-O2174 from 50.1.100.233 of 111186.0 MB, 45.9 cpus for [*]
2018-01-09 07:34:56,692 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            -   59b85b42-a4a2-4632-9578-9e480585ecdc-O2175 from 50.1.100.235 of 123506.0 MB, 47.3 cpus for [*]
2018-01-09 07:34:56,692 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            -   59b85b42-a4a2-4632-9578-9e480585ecdc-O2176 from 50.1.100.234 of 124530.0 MB, 46.6 cpus for [*]
2018-01-09 07:34:56,692 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            -   59b85b42-a4a2-4632-9578-9e480585ecdc-O2177 from 50.1.100.231 of 123474.0 MB, 47.2 cpus for [*]
2018-01-09 07:34:56,693 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            -   59b85b42-a4a2-4632-9578-9e480585ecdc-O2178 from 50.1.100.232 of 123474.0 MB, 47.2 cpus for [*]
2018-01-09 07:34:57,711 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            - Processing 5 task(s) against 5 new offer(s) plus outstanding offers.
2018-01-09 07:34:57,726 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            - Resources considered: (note: expired offers not deducted from below)
2018-01-09 07:34:57,727 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            -   50.1.100.234 has 124530.0 MB, 46.6 cpus
2018-01-09 07:34:57,728 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            -   50.1.100.235 has 123506.0 MB, 47.3 cpus
2018-01-09 07:34:57,728 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            -   50.1.100.232 has 123474.0 MB, 47.2 cpus
2018-01-09 07:34:57,728 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            -   50.1.100.233 has 111186.0 MB, 45.9 cpus
2018-01-09 07:34:57,728 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            -   50.1.100.231 has 123474.0 MB, 47.2 cpus
2018-01-09 07:34:58,069 INFO  org.apache.flink.mesos.runtime.clusterframework.MesosFlinkResourceManager  - Launching Mesos task taskmanager-00005 on host 50.1.100.231.
2018-01-09 07:34:58,069 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            - Launched 5 task(s) on 50.1.100.231 using 1 offer(s):
2018-01-09 07:34:58,070 INFO  org.apache.flink.mesos.runtime.clusterframework.MesosFlinkResourceManager  - Launching Mesos task taskmanager-00002 on host 50.1.100.231.
2018-01-09 07:34:58,070 INFO  org.apache.flink.mesos.runtime.clusterframework.MesosFlinkResourceManager  - Launching Mesos task taskmanager-00003 on host 50.1.100.231.
2018-01-09 07:34:58,070 INFO  org.apache.flink.mesos.runtime.clusterframework.MesosFlinkResourceManager  - Launching Mesos task taskmanager-00004 on host 50.1.100.231.
2018-01-09 07:34:58,070 INFO  org.apache.flink.mesos.runtime.clusterframework.MesosFlinkResourceManager  - Launching Mesos task taskmanager-00001 on host 50.1.100.231.
2018-01-09 07:34:58,070 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            -   59b85b42-a4a2-4632-9578-9e480585ecdc-O2177
2018-01-09 07:34:58,071 INFO  org.apache.flink.mesos.scheduler.LaunchCoordinator            - No longer gathering offers