Page History

Myriad allows Mesos and YARN to co-exist and share resources with Mesos as the resource manager for the data center. Sharing resources between these two resource allocation systems improves overall cluster utilization and avoids statically partitioning resources amongst two among separate clusters and separate resource managers.

The following diagram shows two resource managers running independently which results in a statically partitioned data center:

...

The Mesos Slave and YARN’s Node Manager are processes that run on the host OS. Both processes advertise available resources to Mesos Master and YARN’s Resource Manager respectively. Each process can be configured to advertise a subset of resources. This ability is leveraged, in conjunction with cgroups, to allow Mesos Slave and YARN’s Node Manager to co-exist on a node.

• The Mesos Slave processes advertises all of a node’s resources (8 CPUs, 16 GB RAM) to the Mesos Master.
• The YARN Node Manager is started as a Mesos Task. This task is allotted (4 CPUs and 8 GB RAM) and the Node Manager is configured to only advertise 3 CPUs and 7 GB RAM.

• The Node Manager is also configured to mount the YARN containers under the cgroup hierarchy  which stems from

  a Mesos task.

...

• a Mesos

...

• task.

...

Image Removed

• For example:

  Code Block

...

• /sys/fs/cgroup/cpu/mesos/node-manager-task-id/container-1

The following diagram shows a node running YARN Node Manager as a Mesos Slave task.

Image Added

High Level Design

One way to avoid static partitioning and to enable resource sharing when running two resource managers, is to let one resource manager be in absolute control of the datacenter’s resources. The other resource manager then manages a subset of resources, allocated to it through the primary resource manager.

...