diff --git a/dev/.rat-excludes b/dev/.rat-excludes
index 6c809f4341..a35d4ce976 100644
--- a/dev/.rat-excludes
+++ b/dev/.rat-excludes
@@ -134,4 +134,5 @@ flights_tiny.txt.1
 over1k
 over10k
 exported_table/*
+ansible-for-test-node/*
 node_modules
\ No newline at end of file
diff --git a/dev/ansible-for-test-node/README.md b/dev/ansible-for-test-node/README.md
new file mode 100644
index 0000000000..846bea514a
--- /dev/null
+++ b/dev/ansible-for-test-node/README.md
@@ -0,0 +1,25 @@
+# jenkins-infra
+
+This is a rough skeleton of the ansible used to deploy RISELab/Apache Spark Jenkins build workers on Ubuntu 20LTS.
+
+WARNING:  this will not work "directly out of the box" and will need to be tweaked to work on any ubuntu servers you might want to try this on.
+
+### deploy a new worker node
+#### TL;DR:
+all of the configs for the workers live in roles/common/... and roles/jenkins-worker...
+
+#### prereqs:
+* fresh install of ubuntu 20
+* a service account w/sudo
+* python 3, ansible, ansible-playbook installed locally
+* add hostname(s) to the `hosts` file
+* add this to your `~/.ansible.cfg`:
+```[defaults] host_key_checking = False```
+
+#### fire ansible cannon!
+`ansible-playbook -u <service account> deploy-jenkins-worker.yml -i <ansible-style-hosts-file> -k -b -K` 
+
+tips:
+* if you are installing more than a few workers, it's best to run the playbook on smaller (2-3) batches at a time.  this way it's easier to track down errors, as ansible is very noisy.
+* when you encounter an error, you should comment out any previously-run plays and tasks.  this saves time when debugging, and let's you easily track where you are in the process.
+* `apt-get remove <application_name>` and `apt-get purge <package-name>` are your friends
diff --git a/dev/ansible-for-test-node/deploy-jenkins-worker.yml b/dev/ansible-for-test-node/deploy-jenkins-worker.yml
new file mode 100644
index 0000000000..7ab5f54349
--- /dev/null
+++ b/dev/ansible-for-test-node/deploy-jenkins-worker.yml
@@ -0,0 +1,8 @@
+---
+# top-level setup for jenkins workers
+- name: set up ubuntu jenkins workers
+  hosts: ubuntu-build-node
+  roles:
+    - common
+    - jenkins-worker
+  become: yes
diff --git a/dev/ansible-for-test-node/roles/common/README.md b/dev/ansible-for-test-node/roles/common/README.md
new file mode 100644
index 0000000000..bff1359cd7
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/common/README.md
@@ -0,0 +1,4 @@
+Role Name
+=========
+
+common -- just some common things to set up for any jenkins worker node
diff --git a/dev/ansible-for-test-node/roles/common/tasks/main.yml b/dev/ansible-for-test-node/roles/common/tasks/main.yml
new file mode 100644
index 0000000000..607f5f2875
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/common/tasks/main.yml
@@ -0,0 +1,4 @@
+---
+# tasks file for common
+- include: system_packages.yml
+- include: setup_local_userspace.yml
diff --git a/dev/ansible-for-test-node/roles/common/tasks/setup_local_userspace.yml b/dev/ansible-for-test-node/roles/common/tasks/setup_local_userspace.yml
new file mode 100644
index 0000000000..669700a34c
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/common/tasks/setup_local_userspace.yml
@@ -0,0 +1,8 @@
+---
+# set up userspace
+- name: create local groups for external contributors
+  group:
+    name: "{{ item }}"
+    state: present
+  loop:
+    - jenkins
diff --git a/dev/ansible-for-test-node/roles/common/tasks/system_packages.yml b/dev/ansible-for-test-node/roles/common/tasks/system_packages.yml
new file mode 100644
index 0000000000..e6628987a8
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/common/tasks/system_packages.yml
@@ -0,0 +1,73 @@
+---
+# base system packages for jenkins master and workers
+- name: get software for apt repository management.
+  apt:
+    pkg:
+      - python3-apt
+      - python3-pycurl
+    state: present
+
+- name: install java11
+  apt:
+    pkg: openjdk-11-jdk
+    state: present
+    update_cache: yes
+    force: yes
+
+- name: install java8
+  apt:
+    pkg: openjdk-8-jdk
+    state: present
+    update_cache: yes
+    force: yes
+
+- name: make easy to remember dir for java symlinks
+  file:
+    path: /usr/java
+    state: directory
+
+- name: link java8 to /usr/java/latest
+  file:
+    src: /usr/lib/jvm/java-8-openjdk-amd64/
+    dest: /usr/java/latest
+    state: link
+
+- name: link java8 to /usr/java/java8
+  file:
+    src: /usr/lib/jvm/java-8-openjdk-amd64/
+    dest: /usr/java/java8
+    state: link
+
+- name: link java11 to /usr/java/java11
+  file:
+    src: /usr/lib/jvm/java-11-openjdk-amd64/
+    dest: /usr/java/java11
+    state: link
+
+- name: set java8 as default
+  alternatives:
+    name: "{{ item.command }}"
+    path: "{{ item.path }}"
+  loop:
+    - { command: 'java', path: '/usr/java/latest/bin/java' }
+    - { command: 'javah', path: '/usr/java/latest/bin/javah' }
+    - { command: 'javac', path: '/usr/java/latest/bin/javac' }
+    - { command: 'jar', path: '/usr/java/latest/bin/jar' }
+
+- name: install base system packages
+  apt:
+    pkg:
+      - apt-transport-https
+      - gnupg-agent
+      - ipmitool
+      - git
+      - ntp
+      - htop
+      - tree
+      - gzip
+      - bzip2
+      - curl
+      - wget
+      - unzip
+    state: present
+    update_cache: yes
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/README.md b/dev/ansible-for-test-node/roles/jenkins-worker/README.md
new file mode 100644
index 0000000000..71b0a218c7
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/README.md
@@ -0,0 +1,15 @@
+Role Name
+=========
+
+jenkins-worker -- set up the craziness of a jenkins worker to build and test Apache Spark
+
+Requirements
+------------
+
+Oh jeez.  This is just a framework to help others get started.  If you try and deploy this locally, you'll need a service account, auth set up, etc etc.
+
+Role Variables
+--------------
+
+vars/main.yml:  git caches, and url construction for minikube downloads
+defaults/main.yml:  urls, versions, install targets, etc
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/defaults/main.yml b/dev/ansible-for-test-node/roles/jenkins-worker/defaults/main.yml
new file mode 100644
index 0000000000..98092229ca
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/defaults/main.yml
@@ -0,0 +1,30 @@
+---
+# defaults file for jenkins-worker
+anaconda_installer: https://repo.continuum.io/archive/{{ anaconda_version }}
+anaconda_version: Anaconda2-2019.10-Linux-x86_64.sh
+anaconda_home: /home/jenkins/anaconda2
+anaconda_py3_pip_requirements: base-py3-pip.txt
+
+spark_py36_environment: spark-py36-spec.txt
+spark_py3k_environment: spark-py3k-spec.txt
+spark_py2_pip_requirements: spark-py2-pip.txt
+
+jenkins_home: /home/jenkins
+
+minikube_version: 1.18.1
+minikube_checksum: sha256:1a7960b845301107cb6a0c29001c8df310d7bce586cf88ceacfc78f22b622ba5
+minikube_mirror: https://github.com/kubernetes/minikube/releases/download
+minikube_target: "minikube_{{ minikube_version }}-0_amd64.deb"
+
+k8s_version: 1.17.3
+
+kubectl_version: v1.17.3
+kubectl_mirror: https://dl.k8s.io/release
+kubectl_target: 'kubectl'
+kubectl_install_dir: /usr/local/bin
+
+pypy_pip_module: get-pip.py
+pypy_pip_mirror: https://bootstrap.pypa.io
+
+r_cran_repo: "deb https://cloud.r-project.org/bin/linux/ubuntu bionic-cran35/"
+r_cran_repo_key: E298A3A825C0D65DFD57CBB651716619E084DAB9
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/base-py3-pip.txt b/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/base-py3-pip.txt
new file mode 100644
index 0000000000..af3cbd04e5
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/base-py3-pip.txt
@@ -0,0 +1,3 @@
+awscli==1.16.46
+pytest-benchmark==3.1.1
+pytest-html==1.19.0
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/base-py3-spec.txt b/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/base-py3-spec.txt
new file mode 100644
index 0000000000..91c988017a
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/base-py3-spec.txt
@@ -0,0 +1,21 @@
+# This file may be used to create an environment using:
+# $ conda create --name <env> --file <this file>
+# platform: linux-64
+@EXPLICIT
+https://repo.anaconda.com/pkgs/main/linux-64/ca-certificates-2018.03.07-0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libgcc-ng-8.2.0-hdf63c60_1.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libstdcxx-ng-8.2.0-hdf63c60_1.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libffi-3.2.1-hd88cf55_4.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/ncurses-6.1-hf484d3e_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/openssl-1.0.2p-h14c3975_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/xz-5.2.4-h14c3975_4.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/zlib-1.2.11-ha838bed_2.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libedit-3.1.20170329-h6b74fdf_2.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/readline-7.0-h7b6447c_5.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/tk-8.6.8-hbc83047_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/sqlite-3.24.0-h84994c4_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/python-3.6.6-hc3d631a_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/certifi-2018.8.24-py36_1.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/setuptools-40.2.0-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/wheel-0.31.1-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/pip-10.0.1-py36_0.tar.bz2
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/py36.txt b/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/py36.txt
new file mode 100644
index 0000000000..50b72f41fe
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/py36.txt
@@ -0,0 +1,49 @@
+# This file may be used to create an environment using:
+# $ conda create --name <env> --file <this file>
+# platform: linux-64
+@EXPLICIT
+https://repo.anaconda.com/pkgs/main/linux-64/blas-1.0-mkl.tar.bz2
+https://conda.anaconda.org/anaconda/linux-64/ca-certificates-2020.1.1-0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/intel-openmp-2019.3-199.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libffi-3.2.1-hd88cf55_4.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libgcc-ng-8.2.0-hdf63c60_1.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libgfortran-ng-7.3.0-hdf63c60_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libstdcxx-ng-8.2.0-hdf63c60_1.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/bzip2-1.0.6-h14c3975_1002.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/icu-58.2-hf484d3e_1000.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/mkl-2019.3-199.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/ncurses-6.1-he6710b0_1.tar.bz2
+https://conda.anaconda.org/anaconda/linux-64/openssl-1.1.1-h7b6447c_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/xz-5.2.4-h14c3975_4.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/zlib-1.2.11-h7b6447c_3.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/boost-cpp-1.68.0-h11c811c_1000.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libedit-3.1.20181209-hc058e9b_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/libprotobuf-3.6.1-hdbcaa40_1001.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/readline-7.0-h7b6447c_5.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/tk-8.6.8-hbc83047_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/sqlite-3.27.2-h7b6447c_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/thrift-cpp-0.12.0-h0a07b25_1002.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/python-3.6.8-h0371630_0.tar.bz2
+https://conda.anaconda.org/anaconda/linux-64/certifi-2019.11.28-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/coverage-4.5.2-py36h7b6447c_0.tar.bz2
+https://conda.anaconda.org/anaconda/linux-64/entrypoints-0.3-py36_0.tar.bz2
+https://conda.anaconda.org/anaconda/linux-64/mccabe-0.6.1-py36_1.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/numpy-base-1.16.2-py36hde5b4d6_0.tar.bz2
+https://conda.anaconda.org/anaconda/linux-64/pycodestyle-2.5.0-py36_0.tar.bz2
+https://conda.anaconda.org/anaconda/linux-64/pyflakes-2.1.1-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/pytz-2018.9-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/six-1.12.0-py36_0.tar.bz2
+https://conda.anaconda.org/conda-forge/noarch/xmlrunner-1.7.7-py_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/mkl_random-1.0.2-py36hd81dba3_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/python-dateutil-2.8.0-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/setuptools-40.8.0-py36_0.tar.bz2
+https://conda.anaconda.org/anaconda/linux-64/flake8-3.7.9-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/wheel-0.33.1-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/pip-19.0.3-py36_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/arrow-cpp-0.12.1-py36h0e61e49_0.tar.bz2
+https://conda.anaconda.org/conda-forge/noarch/parquet-cpp-1.5.1-4.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/pandas-0.24.2-py36hf484d3e_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/pyarrow-0.12.1-py36hbbcf98d_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/mkl_fft-1.0.10-py36ha843d7b_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/numpy-1.16.2-py36h7e9f1db_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/scipy-1.2.1-py36h7c811a0_0.tar.bz2
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/spark-py2-pip.txt b/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/spark-py2-pip.txt
new file mode 100644
index 0000000000..74ec2eefd7
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/spark-py2-pip.txt
@@ -0,0 +1,8 @@
+coverage==4.5.2
+dask
+flake8==3.6.0
+numpy
+pandas
+pyarrow==0.8.0
+pycodestyle==2.4.0
+pyflakes==2.0.0
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/spark-py36-spec.txt b/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/spark-py36-spec.txt
new file mode 100644
index 0000000000..029c535831
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/spark-py36-spec.txt
@@ -0,0 +1,61 @@
+# This file may be used to create an environment using:
+# $ conda create --name <env> --file <this file>
+# platform: linux-64
+@EXPLICIT
+https://repo.anaconda.com/pkgs/main/linux-64/_libgcc_mutex-0.1-main.conda
+https://repo.anaconda.com/pkgs/main/linux-64/blas-1.0-mkl.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/ca-certificates-2019.11.28-hecc5488_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/intel-openmp-2019.3-199.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libffi-3.2.1-hd88cf55_4.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libgcc-ng-8.2.0-hdf63c60_1.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libgfortran-ng-7.3.0-hdf63c60_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libstdcxx-ng-8.2.0-hdf63c60_1.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/brotli-1.0.7-he1b5a44_1000.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/bzip2-1.0.8-h516909a_2.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/c-ares-1.15.0-h516909a_1001.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/double-conversion-3.1.5-he1b5a44_2.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/gflags-2.2.2-he1b5a44_1002.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/glog-0.4.0-he1b5a44_1.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/icu-58.2-hf484d3e_1000.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/lz4-c-1.8.3-he1b5a44_1001.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/mkl-2019.3-199.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/ncurses-6.1-he6710b0_1.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/openssl-1.1.1d-h516909a_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/re2-2020.03.03-he1b5a44_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/snappy-1.1.8-he1b5a44_1.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/uriparser-0.9.3-he1b5a44_1.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/xz-5.2.4-h14c3975_4.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/zlib-1.2.11-h7b6447c_3.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/boost-cpp-1.68.0-h11c811c_1000.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libedit-3.1.20181209-hc058e9b_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/libevent-2.1.10-h72c5cf5_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libprotobuf-3.11.2-hd408876_0.conda
+https://repo.anaconda.com/pkgs/main/linux-64/readline-7.0-h7b6447c_5.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/tk-8.6.8-hbc83047_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/zstd-1.3.7-h0b5b093_0.conda
+https://repo.anaconda.com/pkgs/main/linux-64/grpc-cpp-1.26.0-hf8bcb03_0.conda
+https://repo.anaconda.com/pkgs/main/linux-64/libboost-1.71.0-h97c9712_0.conda
+https://repo.anaconda.com/pkgs/main/linux-64/sqlite-3.27.2-h7b6447c_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/thrift-cpp-0.11.0-h02b749d_3.conda
+https://repo.anaconda.com/pkgs/main/linux-64/python-3.6.8-h0371630_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/certifi-2019.11.28-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/coverage-4.5.2-py36h7b6447c_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/mccabe-0.6.1-py36_1.conda
+https://repo.anaconda.com/pkgs/main/linux-64/numpy-base-1.16.2-py36hde5b4d6_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/pycodestyle-2.4.0-py36_0.conda
+https://repo.anaconda.com/pkgs/main/linux-64/pyflakes-2.0.0-py36_0.conda
+https://repo.anaconda.com/pkgs/main/linux-64/pytz-2018.9-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/six-1.12.0-py36_0.tar.bz2
+https://conda.anaconda.org/conda-forge/noarch/xmlrunner-1.7.7-py_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/mkl_random-1.0.2-py36hd81dba3_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/python-dateutil-2.8.0-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/setuptools-40.8.0-py36_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/flake8-3.6.0-py36_1000.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/wheel-0.33.1-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/pip-19.0.3-py36_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/pandas-0.24.2-py36hf484d3e_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/arrow-cpp-0.15.1-py36h7cd5009_5.conda
+https://repo.anaconda.com/pkgs/main/linux-64/mkl_fft-1.0.10-py36ha843d7b_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/numpy-1.16.2-py36h7e9f1db_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/pyarrow-0.15.1-py36h0573a6f_0.conda
+https://repo.anaconda.com/pkgs/main/linux-64/scipy-1.2.1-py36h7c811a0_0.tar.bz2
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/spark-py3k-spec.txt b/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/spark-py3k-spec.txt
new file mode 100644
index 0000000000..3f6770b70c
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/files/python_environments/spark-py3k-spec.txt
@@ -0,0 +1,42 @@
+# This file may be used to create an environment using:
+# $ conda create --name <env> --file <this file>
+# platform: linux-64
+@EXPLICIT
+https://repo.anaconda.com/pkgs/main/linux-64/blas-1.0-mkl.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/ca-certificates-2019.3.9-hecc5488_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/intel-openmp-2019.3-199.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libffi-3.2.1-hd88cf55_4.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libgcc-ng-8.2.0-hdf63c60_1.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libgfortran-ng-7.3.0-hdf63c60_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libstdcxx-ng-8.2.0-hdf63c60_1.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/bzip2-1.0.6-h14c3975_1002.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/icu-58.2-hf484d3e_1000.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/mkl-2019.3-199.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/ncurses-6.1-he6710b0_1.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/openssl-1.1.1b-h14c3975_1.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/xz-5.2.4-h14c3975_4.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/zlib-1.2.11-h7b6447c_3.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/boost-cpp-1.68.0-h11c811c_1000.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/libedit-3.1.20181209-hc058e9b_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/libprotobuf-3.6.1-hdbcaa40_1001.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/readline-7.0-h7b6447c_5.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/tk-8.6.8-hbc83047_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/sqlite-3.27.2-h7b6447c_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/thrift-cpp-0.12.0-h0a07b25_1002.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/python-3.6.8-h0371630_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/certifi-2019.3.9-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/coverage-4.5.2-py36h7b6447c_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/numpy-1.11.3-py36h3dfced4_4.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/pytz-2018.9-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/six-1.12.0-py36_0.tar.bz2
+https://conda.anaconda.org/conda-forge/noarch/xmlrunner-1.7.7-py_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/arrow-cpp-0.8.0-py36_4.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/mkl_fft-1.0.11-py36h14c3975_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/mkl_random-1.0.2-py36h637b7d7_2.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/python-dateutil-2.8.0-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/setuptools-40.8.0-py36_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/pandas-0.19.2-np111py36_1.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/parquet-cpp-1.4.0.pre-2.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/wheel-0.33.1-py36_0.tar.bz2
+https://repo.anaconda.com/pkgs/main/linux-64/pip-19.0.3-py36_0.tar.bz2
+https://conda.anaconda.org/conda-forge/linux-64/pyarrow-0.8.0-py36_0.tar.bz2
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/files/scripts/jenkins-gitcache-cron b/dev/ansible-for-test-node/roles/jenkins-worker/files/scripts/jenkins-gitcache-cron
new file mode 100644
index 0000000000..482c86c930
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/files/scripts/jenkins-gitcache-cron
@@ -0,0 +1,7 @@
+# this must be run as the jenkins user!
+SHELL=/bin/bash
+PATH=/home/jenkins/git2/bin:/usr/local/bin:/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/sbin
+
+*/5 * * * *     cd /home/jenkins/gitcaches/adam.reference && git remote update >/dev/null 2>&1
+*/5 * * * *     cd /home/jenkins/gitcaches/alluxio.reference && git remote update >/dev/null 2>&1
+*/5 * * * *     cd /home/jenkins/gitcaches/spark.reference && git remote update >/dev/null 2>&1
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/files/util_scripts/kill_zinc_nailgun.py b/dev/ansible-for-test-node/roles/jenkins-worker/files/util_scripts/kill_zinc_nailgun.py
new file mode 100755
index 0000000000..40887e8977
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/files/util_scripts/kill_zinc_nailgun.py
@@ -0,0 +1,60 @@
+#!/usr/bin/env python3
+"""Kill a Zinc process that is listening on a given port"""
+import argparse
+import os
+import re
+import signal
+import subprocess
+import sys
+
+
+def _parse_args():
+    zinc_port_var = "ZINC_PORT"
+    zinc_port_option = "--zinc-port"
+    parser = argparse.ArgumentParser()
+    parser.add_argument(zinc_port_option,
+                        type=int,
+                        default=int(os.environ.get(zinc_port_var, "0")),
+                        help="Specify zinc port")
+    args = parser.parse_args()
+    if not args.zinc_port:
+        parser.error("Specify either environment variable {0} or option {1}".format(
+            zinc_port_var, zinc_port_option))
+    return args
+
+
+def _kill_processes_listening_on_port(port):
+    killed = set()
+    for pid in _yield_processes_listening_on_port(port):
+        if not pid in killed:
+            killed.add(pid)
+            os.kill(pid, signal.SIGTERM)
+
+
+def _yield_processes_listening_on_port(port):
+    pattern = re.compile(r":{0} \(LISTEN\)".format(port))
+    innocuous_errors = re.compile(
+        r"^\s*Output information may be incomplete.\s*$"
+        r"|^lsof: WARNING: can't stat\(\) (?:tracefs|nsfs|overlay|tmpfs|aufs|zfs) file system .*$"
+        r"|^\s*$")
+    lsof_process = subprocess.Popen(["lsof", "-P"], stdout=subprocess.PIPE,
+                                    stderr=subprocess.PIPE, universal_newlines=True)
+    stdout, stderr = lsof_process.communicate()
+    if lsof_process.returncode != 0:
+        raise OSError("Can't run lsof -P, stderr:\n{}".format(stderr))
+    for line in stderr.split("\n"):
+        if not innocuous_errors.match(line):
+            sys.stderr.write(line + "\n")
+    for line in stdout.split("\n"):
+        if pattern.search(line):
+            yield int(line.split()[1])
+
+
+def _main():
+    args = _parse_args()
+    _kill_processes_listening_on_port(args.zinc_port)
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(_main())
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/files/util_scripts/post_github_pr_comment.py b/dev/ansible-for-test-node/roles/jenkins-worker/files/util_scripts/post_github_pr_comment.py
new file mode 100755
index 0000000000..68e31d4528
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/files/util_scripts/post_github_pr_comment.py
@@ -0,0 +1,81 @@
+#!/usr/bin/env python3
+"""Utility program to post a comment to a github PR"""
+import argparse
+import json
+import os
+import sys
+import urllib.parse
+from urllib.error import HTTPError, URLError
+from urllib.request import urlopen, Request
+
+
+def _parse_args():
+    pr_link_var = "ghprbPullLink"
+    pr_link_option = "--pr-link"
+    github_oauth_key_var = "GITHUB_OAUTH_KEY"
+    github_oauth_key_option = "--github-oauth-key"
+    parser = argparse.ArgumentParser()
+    parser.add_argument("-pr", pr_link_option,
+                        default=os.environ.get(pr_link_var, ""),
+                        help="Specify pull request link")
+    parser.add_argument(github_oauth_key_option,
+                        default=os.environ.get(github_oauth_key_var, ""),
+                        help="Specify github oauth key")
+    args = parser.parse_args()
+    if not args.pr_link:
+        parser.error("Specify either environment variable {} or option {}".format(
+            pr_link_var, pr_link_option))
+    if not args.github_oauth_key:
+        parser.error("Specify either environment variable {} or option {}".format(
+            github_oauth_key_var, github_oauth_key_option))
+    return args
+
+
+def post_message_to_github(msg, github_oauth_key, pr_link):
+    print("Attempting to post to Github...")
+
+    ghprb_pull_id = os.environ["ghprbPullId"]
+    api_url = os.getenv("GITHUB_API_BASE", "https://api.github.com/repos/apache/spark")
+    url = api_url + "/issues/" + ghprb_pull_id + "/comments"
+
+    posted_message = json.dumps({"body": msg})
+    request = Request(url,
+                      headers={
+                          "Authorization": "token {}".format(github_oauth_key),
+                          "Content-Type": "application/json"
+                      },
+                      data=posted_message.encode('utf-8'))
+    try:
+        response = urlopen(request)
+
+        if response.getcode() == 201:
+            print(" > Post successful.")
+        else:
+            print_err("Surprising post response.")
+            print_err(" > http_code: {}".format(response.getcode()))
+            print_err(" > api_response: {}".format(response.read()))
+            print_err(" > data: {}".format(posted_message))
+    except HTTPError as http_e:
+        print_err("Failed to post message to Github.")
+        print_err(" > http_code: {}".format(http_e.code))
+        print_err(" > api_response: {}".format(http_e.read()))
+        print_err(" > data: {}".format(posted_message))
+    except URLError as url_e:
+        print_err("Failed to post message to Github.")
+        print_err(" > urllib_status: {}".format(url_e.reason[1]))
+        print_err(" > data: {}".format(posted_message))
+
+
+def print_err(msg):
+    print(msg, file=sys.stderr)
+
+
+def _main():
+    args = _parse_args()
+    msg = sys.stdin.read()
+    post_message_to_github(msg, args.github_oauth_key, args.pr_link)
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(_main())
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/files/util_scripts/session_lock_resource.py b/dev/ansible-for-test-node/roles/jenkins-worker/files/util_scripts/session_lock_resource.py
new file mode 100755
index 0000000000..f5153d5161
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/files/util_scripts/session_lock_resource.py
@@ -0,0 +1,152 @@
+#!/usr/bin/env python2
+"""Utility program to hold resources for a given session.
+
+A session is defined as the lifetime of some process (externally specified,
+or the original forking process of this process)
+
+This program will daemonize and return 0 if the lock succeeds. If the lock
+fails, it will return a non-zero exit code.
+
+After the parent program (or other specified pid) exits, it will also exit,
+unlocking the file
+"""
+import argparse
+import errno
+import fcntl
+import os
+import sys
+import time
+
+
+_LOCK_DIR = "/tmp/session_locked_resources"
+
+
+def _parse_args():
+    """Parse command line arguments"""
+    parser = argparse.ArgumentParser()
+    parser.add_argument("-t", "--timeout-secs", type=int,
+                        help="How long to wait for lock acquisition, in seconds")
+    parser.add_argument("-p", "--pid", type=int,
+                        help="PID to wait for exit (defaults to parent pid)")
+    parser.add_argument("resource", help="Resource to lock")
+    return parser.parse_args()
+
+
+def _acquire_lock(filename, timeout_secs, message):
+    """Acquire a lock file.
+
+    Returns True iff the file could be locked within the timeout
+    """
+    f = open(filename, "a+")
+    time_attempted = 0
+    while True:
+        try:
+            fcntl.lockf(f.fileno(), fcntl.LOCK_EX | fcntl.LOCK_NB)
+            # Lock acquired
+            f.truncate(0)
+            f.write(message)
+            f.flush()
+            return f
+        except IOError:
+            # Locking failed
+            time_attempted += 1
+            if timeout_secs and time_attempted >= timeout_secs:
+                # Timeout exceeded
+                raise IOError("Can't get child lock")
+        time.sleep(1)
+
+
+def _daemonize(child_body):
+    """Daemonize. Returns whether the child was successful.
+
+    Child body is a function to call in the child. It should take one
+    argument, which is a function that is called with a boolean
+    that indicates that the child succeeded/failed in its initialization
+    """
+    CHILD_FAIL = '\2'
+    CHILD_SUCCESS = '\0'
+    r_fd, w_fd = os.pipe()
+    if os.fork() != 0:
+        # We are the original script. Read success/fail from the final
+        # child and log an error message if needed.
+        child_code = os.read(r_fd, 1) # .decode('utf-8')
+        return child_code == CHILD_SUCCESS
+    # First child
+    os.setsid()
+    if os.fork() != 0:
+        # Still in first child
+        _close_std_streams()
+        os._exit(0)
+    # Second child (daemon process)
+
+    def _write_to_parent(success):
+        parent_message = CHILD_SUCCESS if success else CHILD_FAIL
+        os.write(w_fd, parent_message)
+    child_body(_write_to_parent)
+    os._exit(0)
+
+
+def _close_std_streams():
+    """Close all our stdin/stdout/stderr streams."""
+    sys.stdin.close()
+    sys.stdout.close()
+    sys.stderr.close()
+    os.close(0)
+    os.close(1)
+    os.close(2)
+
+
+def _wait_for_pid_exit(pid):
+    while _is_pid_running(pid):
+        time.sleep(1)
+
+
+def _is_pid_running(pid):
+    """Wait for a pid to finish.
+
+    From Stack Overflow: https://stackoverflow.com/questions/7653178
+    """
+    try:
+        os.kill(pid, 0)
+    except OSError as err:
+        if err.errno == errno.ESRCH:
+            return False
+    return True
+
+
+def _lock_and_wait(lock_success_callback, resource, timeout_secs,
+                   controlling_pid):
+    """Attempt to lock the file then wait.
+
+    lock_success_callback will be called if the locking worked.
+    """
+    lock_filename = os.path.join(_LOCK_DIR, resource)
+    lock_message = ("Session lock on " + resource
+        + ", controlling pid " + str(controlling_pid) + "\n")
+    try:
+        f = _acquire_lock(lock_filename, timeout_secs, lock_message)
+    except IOError:
+        lock_success_callback(False)
+        return
+    lock_success_callback(True)
+    _wait_for_pid_exit(controlling_pid)
+
+
+def main():
+    """Main program"""
+    args = _parse_args()
+    if not os.path.exists(_LOCK_DIR):
+        os.mkdir(_LOCK_DIR)
+    controlling_pid = args.pid or os.getppid()
+    child_body_func = lambda success_callback: _lock_and_wait(
+        success_callback, args.resource, args.timeout_secs,
+        controlling_pid)
+    if _daemonize(child_body_func):
+        return 0
+    else:
+        print("Could not acquire lock")
+        return 1
+
+
+if __name__ == "__main__":
+    sys.exit(main())
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/files/worker-limits.conf b/dev/ansible-for-test-node/roles/jenkins-worker/files/worker-limits.conf
new file mode 100644
index 0000000000..1a9901a636
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/files/worker-limits.conf
@@ -0,0 +1,5 @@
+*          soft    nproc     1024
+root       soft    nproc     unlimited
+jenkins        soft    nofile    100000
+jenkins        hard    nofile    200000
+jenkins       soft    nproc     1600000
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/tasks/cleanup.yml b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/cleanup.yml
new file mode 100644
index 0000000000..ec700ce091
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/cleanup.yml
@@ -0,0 +1,12 @@
+- name: purge unneeded dependencies
+  apt:
+    purge: yes
+    autoremove: yes
+
+- name: clean up unneeded cached packages
+  apt:
+    autoclean: yes
+
+- name: reset apt to a useful state
+  apt:
+    update_cache: yes
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_anaconda.yml b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_anaconda.yml
new file mode 100644
index 0000000000..7a90431cc6
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_anaconda.yml
@@ -0,0 +1,79 @@
+---
+- name: Check for Anaconda
+  stat:
+    path: "{{ anaconda_home }}"
+  register: anaconda_dir
+
+- block:
+  - name: Download Anaconda
+    get_url:
+      url: "{{ anaconda_installer }}"
+      dest: /tmp/{{ anaconda_version }}
+
+  - name: Install Anaconda
+    command: "bash /tmp/{{ anaconda_version }} -b -p {{ anaconda_home }}"
+    become: yes
+    become_user: jenkins
+
+  - name: add anaconda bin to jenkins PATH in bashrc
+    lineinfile:
+      name: /home/jenkins/.bashrc
+      state: present
+      insertafter: EOF
+      line: "export PATH=/home/jenkins/anaconda2/bin:$PATH"
+
+  - name: delete Anaconda download
+    file:
+      path: "/tmp/{{ anaconda_version }}"
+      state: absent
+  when: anaconda_dir.stat.islnk is not defined
+
+- name: check for py3 env
+  command: "conda env list"
+  environment:
+    PATH: "{{ anaconda_home }}/bin:{{ ansible_env.PATH }}"
+  changed_when: False
+  check_mode: no
+  register: py3_check
+
+- block:
+  - name: Create Python Env
+    command: "{{ anaconda_home }}/bin/conda create -y --name py3 python=3.6"
+    become: yes
+    become_user: jenkins
+  when: "'py3' not in py3_check.stdout"
+
+- name: update anaconda pip for py2.7
+  command: "{{ anaconda_home }}/bin/pip install --upgrade pip"
+  become: yes
+  become_user: jenkins
+  tags:
+    - skip_ansible_lint
+
+- name: update anaconda pip for py3
+  command: "{{ anaconda_home }}/envs/py3/bin/pip install --upgrade pip"
+  become: yes
+  become_user: jenkins
+  tags:
+    - skip_ansible_lint
+
+- name: copy pip requirments for py3
+  copy:
+    src: "python_environments/{{ anaconda_py3_pip_requirements }}"
+    dest: "/tmp/{{ anaconda_py3_pip_requirements }}"
+    owner: jenkins
+    group: jenkins
+    mode: 0660
+
+- name: install py3 pip packages
+  pip:
+    requirements: "/tmp/{{ anaconda_py3_pip_requirements }}"
+  environment:
+    PATH: "{{ anaconda_home }}/envs/py3/bin:{{ ansible_env.PATH }}"
+  become: yes
+  become_user: jenkins
+
+- name: delete py3 pip requirements file
+  file:
+    path: "/tmp/{{ anaconda_py3_pip_requirements }}"
+    state: absent
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_build_packages.yml b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_build_packages.yml
new file mode 100644
index 0000000000..e55ab06549
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_build_packages.yml
@@ -0,0 +1,21 @@
+---
+# set up the required system packages for a host
+- name: install base build/compilation tools
+  apt:
+    pkg:
+      - autoconf
+      - maven
+      - gcc
+      - make
+      - cmake
+      - g++
+      - libssl-dev
+      - libev-dev
+      - libevent-dev
+      - libffi-dev
+      - curl
+      - python-dev
+      - clang-format
+      - pkg-config
+    state: present
+    update_cache: yes
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_docker.yml b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_docker.yml
new file mode 100644
index 0000000000..eacc56e253
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_docker.yml
@@ -0,0 +1,33 @@
+---
+# set up and install docker
+- name: docker repo signing key
+  apt_key:
+    url: "https://download.docker.com/linux/ubuntu/gpg"
+    state: present
+
+- name: docker repo
+  apt_repository:
+    repo: 'deb [arch=amd64] https://download.docker.com/linux/ubuntu bionic stable'
+    state: present
+    update_cache: yes
+    filename: docker
+
+- name: install docker-ce and docker-compose
+  apt:
+    pkg:
+      - docker-ce
+      - docker-ce-cli
+      - containerd.io
+      - docker-compose
+    state: present
+    update_cache: yes
+
+- name: enable docker service
+  service:
+    name: docker
+    enabled: yes
+
+- name: put jenkins in docker group
+  user:
+    name: jenkins
+    groups: docker
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_minikube.yml b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_minikube.yml
new file mode 100644
index 0000000000..7173df46b5
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_minikube.yml
@@ -0,0 +1,16 @@
+---
+# install minikube
+- name: download kubectl binary
+  get_url:
+    url: '{{ kubectl_url }}'
+    dest: '{{ kubectl_install_dir }}/{{ kubectl_target }}'
+    mode: 0755
+
+- name: install minikube
+  apt:
+    deb: '{{ minikube_url }}'
+
+- name: set k8s version
+  command: "/usr/bin/minikube config set kubernetes-version {{ k8s_version }}"
+  become: yes
+  become_user: jenkins
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_spark_build_packages.yml b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_spark_build_packages.yml
new file mode 100644
index 0000000000..663954fde7
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/install_spark_build_packages.yml
@@ -0,0 +1,183 @@
+---
+# install spark build packages
+- name: install pypy3
+  apt:
+    pkg:
+      - pypy3
+    state: present
+    update_cache: yes
+
+- name: download pip module for pypy
+  get_url:
+    url: '{{ pypy_pip_mirror }}/{{ pypy_pip_module }}'
+    dest: '/tmp/{{ pypy_pip_module }}'
+    mode: 0644
+
+- name: install pip module for pypy
+  command: 'pypy3  /tmp/{{ pypy_pip_module }}'
+  changed_when: False
+
+- name: remove temporary pip module for pypy
+  file:
+    path: '/tmp/{{ pypy_pip_module }}'
+    state: absent
+
+- name: install test coverage package for pypy
+  command: 'pypy3 -m pip install coverage==4.5.2'
+  changed_when: False
+
+- name: check for spark py3k env
+  command: "conda env list"
+  environment:
+    PATH: "{{ anaconda_home }}/bin:{{ ansible_env.PATH }}"
+  changed_when: False
+  check_mode: no
+  register: py3k_check
+
+- block:
+  - name: copy python 3.6 conda environment for spark testing (2.4)
+    copy:
+      src: "python_environments/{{ spark_py3k_environment }}"
+      dest: "/tmp/{{ spark_py3k_environment }}"
+      owner: jenkins
+      group: jenkins
+      mode: 0660
+
+  - name: create py3k conda environment for spark
+    command: "conda create -y --name py3k --file /tmp/{{ spark_py3k_environment }}"
+    environment:
+      PATH: "{{ anaconda_home }}/bin:{{ ansible_env.PATH }}"
+    become: yes
+    become_user: jenkins
+
+  - name: delete py3k conda environment file
+    file:
+      path: "/tmp/{{ spark_py3k_environment }}"
+      state: absent
+  when: "'py3k' not in py3k_check.stdout"
+
+- block:
+  - name: copy python 3.6 conda environment for spark testing (master, 3.x+)
+    copy:
+      src: "python_environments/{{ spark_py36_environment }}"
+      dest: "/tmp/{{ spark_py36_environment }}"
+      owner: jenkins
+      group: jenkins
+      mode: 0660
+
+  - name: create py36 conda environment for spark
+    command: "conda create -y --name py36 --file /tmp/{{ spark_py36_environment }}"
+    environment:
+      PATH: "{{ anaconda_home }}/bin:{{ ansible_env.PATH }}"
+    become: yes
+    become_user: jenkins
+
+  - name: delete py36 conda environment file
+    file:
+      path: "/tmp/{{ spark_py36_environment }}"
+      state: absent
+  when: "'py36' not in py3k_check.stdout"
+
+
+- name: create symlink for anaconda homedir
+  file:
+    src: "{{ anaconda_home }}"
+    dest: "/home/anaconda"
+    state: link
+
+- name: copy python2 pip requirements file
+  copy:
+    src: "python_environments/{{ spark_py2_pip_requirements }}"
+    dest: "/tmp/{{ spark_py2_pip_requirements }}"
+    owner: jenkins
+    group: jenkins
+    mode: 0660
+
+- name: install python2 doc and test packages for spark
+  pip:
+    requirements: "/tmp/{{ spark_py2_pip_requirements }}"
+    extra_args: --ignore-installed
+  environment:
+    PATH: "{{ anaconda_home }}/bin:{{ ansible_env.PATH }}"
+  become: yes
+  become_user: jenkins
+
+- name: delete python2 pip requirements file
+  file:
+    path: "/tmp/{{ spark_py2_pip_requirements }}"
+    state: absent
+
+- name: add cran apt repo key
+  apt_key:
+    keyserver: keyserver.ubuntu.com
+    id: "{{ r_cran_repo_key }}"
+    state: present
+
+- name: add cran apt repo
+  apt_repository:
+    repo: "{{ r_cran_repo }}"
+    update_cache: yes
+
+- name: install sparkR system deps
+  apt:
+    pkg:
+      - default-jre
+      - default-jre-headless # both for r-cran-rjava
+      - libcurl4-openssl-dev  # devtools package dep
+      - libxml2-dev  # lintr dep
+      - qpdf
+      - libssh2-1-dev  # devtools dep
+      - libssl-dev # git2r dep
+      - libodbc1
+    state: present
+
+- name: install base R packages
+  apt:
+    pkg:
+      - r-base-core
+      - r-base-dev
+    state: present
+
+- name: ensure R java environment is properly set up
+  command: "/usr/bin/R CMD javareconf"
+  environment:
+    JAVA_HOME: "/usr/java/latest"
+  register: r_result
+  changed_when: False
+  failed_when: "r_result.rc != 0 or 'had non-zero exit status' in r_result.stderr"
+
+- name: install rJava
+  apt:
+    name: r-cran-rjava
+    state: present
+
+- name: install next set java R packages
+  apt:
+    pkg:
+      - r-cran-rodbc
+      - r-mathlib
+
+- name: install required R packages via Rscript (default version)
+  command: /usr/bin/Rscript --slave --no-save --no-restore-history -e "if (! ('{{ item }}' %in% installed.packages()[,'Package'])) { install.packages(pkgs='{{ item }}'); print('Added'); } else { print('Already installed'); }"
+  register: r_result
+  failed_when: "r_result.rc != 0 or 'had non-zero exit status' in r_result.stderr"
+  changed_when: "'Added' in r_result.stdout"
+  loop:
+    - digest
+    - knitr
+    - devtools
+    - plyr
+    - roxygen2
+    - rmarkdown
+    - e1071
+    - testthat
+
+- name: get list of installed packages
+  command: /usr/bin/Rscript -e "installed.packages()[,'Package']"
+  changed_when: False
+  check_mode: no
+  register: r_check
+
+- name: install lintr v2.0.0
+  command: /usr/bin/Rscript --slave --no-save --no-restore-history -e "devtools::install_github('jimhester/lintr@v2.0.0')"
+  when: "'lintr' not in r_check.stdout"
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/tasks/jenkins_userspace.yml b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/jenkins_userspace.yml
new file mode 100644
index 0000000000..04438e1963
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/jenkins_userspace.yml
@@ -0,0 +1,119 @@
+---
+# set up the jenkins worker userspace
+- name: ensure groups jenkins will need are present
+  group:
+    name: "{{ item }}"
+    state: present
+  loop:
+    - jenkins
+    - docker
+
+- name: create jenkins login
+  user:
+    name: jenkins
+    group: jenkins
+    groups: docker
+    generate_ssh_key: yes
+    ssh_key_bits: 2048
+    ssh_key_file: ".ssh/id_rsa"
+    shell: /bin/bash
+
+- name: create the jenkins workspace directory
+  file:
+    path: "{{ jenkins_home }}/workspace"
+    state: directory
+    mode: 0755
+    owner: jenkins
+    group: jenkins
+
+- name: create the jenkins worker bin directory
+  file:
+    path: "{{ jenkins_home }}/bin"
+    state: directory
+    mode: 0755
+    owner: jenkins
+    group: jenkins
+
+- name: copy utility scripts to jenkins bin dir
+  copy:
+    src: "util_scripts/{{ item }}"
+    dest: "{{ jenkins_home }}/bin/"
+    mode: 0755
+    owner: jenkins
+    group: jenkins
+  loop:
+    - session_lock_resource.py
+    - kill_zinc_nailgun.py
+    - post_github_pr_comment.py
+
+- name: update the jenkins proc and open file limits
+  copy:
+    src: worker-limits.conf
+    dest: /etc/security/limits.conf
+    owner: root
+    group: root
+    mode: 0644
+
+- name: create per-executor ivy caches for spark builds
+  file:
+    path: "{{ jenkins_home }}/sparkivy"
+    state: directory
+    mode: 0755
+    owner: jenkins
+    group: jenkins
+
+- name: create the jenkins git cache directory
+  file:
+    path: "{{ jenkins_home }}/gitcaches"
+    state: directory
+    mode: 0755
+    owner: jenkins
+    group: jenkins
+
+- name: create specific local gitcaches for some projects to speed clone time
+  git:
+    accept_hostkey: yes
+    bare: yes
+    dest: "{{ item.value.dir }}"
+    repo: "{{ item.value.repo }}"
+    update: no
+  become: yes
+  become_user: jenkins
+  loop: "{{ lookup('dict', git_caches) }}"
+  tags:
+    - skip_ansible_lint
+
+- name: configure git cache repos
+  git_config:
+    name: gc.pruneexpire
+    value: never
+    repo: "{{ item.value.dir }}"
+    scope: local
+  become: yes
+  become_user: jenkins
+  loop: "{{ lookup('dict', git_caches) }}"
+  tags:
+    - skip_ansible_lint
+
+- name: perform initial fetch
+  shell:
+    "cd {{ item.value.dir }} && /usr/bin/git fetch --all"
+  become: yes
+  become_user: jenkins
+  loop: "{{ lookup('dict', git_caches) }}"
+  changed_when: False
+  tags:
+    - skip_ansible_lint
+
+- name: set up cron job as jenkins to update gitcaches
+  cron:
+    name: "{{ item.name }}"
+    minute: "*/5"
+    job: "{{ item.cmd }}"
+    user: jenkins
+  loop:
+    - {
+    name: "Spark reference update",
+    cmd: "cd /home/jenkins/gitcaches/spark.reference && git remote update >/dev/null 2>&1"
+    }
+
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/tasks/main.yml b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/main.yml
new file mode 100644
index 0000000000..a673bdd626
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/tasks/main.yml
@@ -0,0 +1,22 @@
+---
+# set up the jenkins user first, before anything is installed and set up
+- name: set up the jenkins environment
+  include: jenkins_userspace.yml
+
+- name: install system packages for software builds
+  include: install_build_packages.yml
+
+- name: install docker
+  include: install_docker.yml
+
+- name: install minikube
+  include: install_minikube.yml
+
+- name: install anaconda and associated packages
+  include: install_anaconda.yml
+
+- name: install spark build packages
+  include: install_spark_build_packages.yml
+
+- name: clean up apt
+  include: cleanup.yml
diff --git a/dev/ansible-for-test-node/roles/jenkins-worker/vars/main.yml b/dev/ansible-for-test-node/roles/jenkins-worker/vars/main.yml
new file mode 100644
index 0000000000..7aa3d25066
--- /dev/null
+++ b/dev/ansible-for-test-node/roles/jenkins-worker/vars/main.yml
@@ -0,0 +1,9 @@
+---
+minikube_url: '{{ minikube_mirror }}/v{{ minikube_version }}/{{ minikube_target }}'
+
+kubectl_url: '{{ kubectl_mirror }}/{{ kubectl_version }}/bin/linux/amd64/{{ kubectl_target }}'
+
+git_caches:
+  spark:
+    dir: '{{ jenkins_home }}/gitcaches/spark.reference'
+    repo: 'https://github.com/apache/spark.git'
diff --git a/dev/tox.ini b/dev/tox.ini
index a0f04ef0cb..e1a4cf5ce8 100644
--- a/dev/tox.ini
+++ b/dev/tox.ini
@@ -16,11 +16,11 @@
 [pycodestyle]
 ignore=E203,E226,E241,E305,E402,E722,E731,E741,W503,W504
 max-line-length=100
-exclude=*/target/*,python/pyspark/cloudpickle/*.py,shared.py,python/docs/source/conf.py,work/*/*.py,python/.eggs/*,dist/*,.git/*
+exclude=*/target/*,python/pyspark/cloudpickle/*.py,shared.py,python/docs/source/conf.py,work/*/*.py,python/.eggs/*,dist/*,.git/*,dev/ansible-for-test-node/*
 
 [flake8]
 select = E901,E999,F821,F822,F823,F401,F405,B006
 # Ignore F821 for plot documents in pandas API on Spark.
 ignore = F821
-exclude = python/docs/build/html/*,*/target/*,python/pyspark/cloudpickle/*.py,shared.py*,python/docs/source/conf.py,work/*/*.py,python/.eggs/*,dist/*,.git/*,python/out,python/pyspark/sql/pandas/functions.pyi,python/pyspark/sql/column.pyi,python/pyspark/worker.pyi,python/pyspark/java_gateway.pyi
+exclude = python/docs/build/html/*,*/target/*,python/pyspark/cloudpickle/*.py,shared.py*,python/docs/source/conf.py,work/*/*.py,python/.eggs/*,dist/*,.git/*,python/out,python/pyspark/sql/pandas/functions.pyi,python/pyspark/sql/column.pyi,python/pyspark/worker.pyi,python/pyspark/java_gateway.pyi,dev/ansible-for-test-node/roles/jenkins-worker/files/util_scripts/*.py
 max-line-length = 100