Add configuration for NCSU's Hazel cluster

With thanks to @jtande!

docs/endpoints/configs/ncsu-hazel.yaml

@@ -0,0 +1,27 @@
+display_name: Hazel HTEX
+engine:
+  type: GlobusComputeEngine
+
+  address:
+    type: address_by_interface
+    ifname: ens5f0
+
+  provider:
+    type: LSFProvider
+    launcher:
+      type: SingleNodeLauncher
+
+    walltime: 00:30:00
+    nodes_per_block: 1
+    init_blocks: 1
+    max_blocks: 1
+
+    queue: standard
+
+    worker_init: |
+      module load conda
+      conda activate /path/to/your/condaenv/conda_env
+
+    scheduler_options: "# BSUB -R span[hosts=1]"
+    cores_per_node: 12
+    cmd_timeout: 60

docs/endpoints/endpoint_examples.rst

@@ -206,6 +206,17 @@ scheduler, and uses the ``SrunLauncher`` to launch workers.
 .. literalinclude:: configs/faster.yaml
    :language: yaml
 
+
+Hazel (NCSU)
+^^^^^^^^^^^^
+
+The Hazel cluster uses the LSF scheduler; the following configuration assumes the user
+is running on a login node, uses the |LSFProvider|_ and the |SingleNodeLauncher|_.
+
+.. literalinclude:: configs/ncsu-hazel.yaml
+   :language: yaml
+
+
 Pinning Workers to devices
 ^^^^^^^^^^^^^^^^^^^^^^^^^^
 
@@ -233,5 +244,12 @@ specific identity assigned: ``CUDA_VISIBLE_DEVICES``, ``ROCR_VISIBLE_DEVICES``,
 .. |HighThroughputExecutor| replace:: ``HighThroughputExecutor``
 .. _HighThroughputExecutor: https://parsl.readthedocs.io/en/stable/stubs/parsl.executors.HighThroughputExecutor.html
 
+.. |Providers| replace:: ``Providers``
+.. _Providers: https://parsl.readthedocs.io/en/stable/reference.html#providers
+.. |LSFProvider| replace:: ``LSFProvider``
+.. _LSFProvider: https://parsl.readthedocs.io/en/latest/stubs/parsl.providers.LSFProvider.html
+.. |SingleNodeLauncher| replace:: ``SingleNodeLauncher``
+.. _SingleNodeLauncher: https://parsl.readthedocs.io/en/latest/stubs/parsl.launchers.SingleNodeLauncher.html
+
 .. |GlobusMPIEngine| replace:: ``GlobusMPIEngine``
-.. _GlobusMPIEngine: reference/mpi_engine.html
+.. _GlobusMPIEngine: reference/mpi_engine.html