add LSF scheduler (#588)

Summary:
I prototyped the LSF scheduler for torchx. It supports native, Docker, and Singularity as runtime with a shared filesystem at this moment. I confirmed it worked with Gloo and NCCL on small VPC V100 clusters.

Note: `torchx log` command is available only when the torchx host shares the filesystem with cluster nodes (e.g., NFS).

In a nutshell, the LSF scheduler translates a torchx request to be LSF job submissions (i.e., `bsub`). For distributed apps, it creates multiple `bsub`. I also added lsf to scripts/component_integration_tests.py. Here is the log output with my three-node LSF cluster and you can find dryrun results there.

[component_integration_tests.lsf.txt](https://github.com/pytorch/torchx/files/9424891/component_integration_tests.lsf.txt)

Regarding Singularity image compatibility, it already automates to convert docker images into singularity image format, and so, only we have to do is to generate singularity-exec arguments from torchx requests. Note that users still need to set prefix docker:// for image names if they want to use docker images.

The following are example commands.

**Example: native hello_world and CLI utils**

```
$ torchx run -s lsf -cfg jobdir=/mnt/data/torchx,runtime=native utils.echo --msg hello_world --num_replicas 3
lsf://torchx/echo-pxc3gn5ct061k
$ torchx list -s lsf
$ torchx status lsf://torchx/echo-pxc3gn5ct061k
$ torchx cancel lsf://torchx/echo-pxc3gn5ct061k
$ torchx log --stream stdout lsf://torchx/echo-pxc3gn5ct061k/echo/0
```

**Example: Docker hello_world**
```
$ torchx run -s lsf -cfg jobdir=/mnt/data/torchx,runtime=docker utils.echo --image alpine:latest --msg hello_world --num_replicas 3
```

**Example: Singularity hello_world**
```
$ torchx run -s lsf -cfg jobdir=/mnt/data/torchx,runtime=singularity utils.echo --image docker://alpine:latest --msg hello_world --num_replicas 3
```

**Example: Docker Distributed**
```
$ cp scripts/dist_app.py /mnt/data/dist/
$ torchx run -s lsf -cfg "jobdir=/mnt/data/torchx,runtime=docker,host_network=True" dist.ddp -j 2x2 --gpu 2 --script /data/dist_app.py --mount "type=bind,src=/mnt/data/dist,dst=/data"
```

**Example: Singularity Distributed**
```
$ cp scripts/dist_app.py /mnt/data/dist/
$ torchx run -s lsf -cfg "jobdir=/mnt/data/torchx,runtime=singularity,host_network=True" dist.ddp --image docker://ghcr.io/pytorch/torchx:0.3.0dev0 -j 2x2 --gpu 2 --script /data/dist_app.py --mount "type=bind,src=/mnt/data/dist,dst=/data"
```

Pull Request resolved: #588

Reviewed By: msaroufim

Differential Revision: D40184939

Pulled By: msaroufim

fbshipit-source-id: 5a13d2ee88b3b5cf1b8e5a3f6786b955d47f21f8

Author: takeshi-yoshimura

docs/source/index.rst

@@ -74,6 +74,7 @@ Works With
    schedulers/slurm
    schedulers/ray
    schedulers/aws_batch
+   schedulers/lsf
 
 .. fbcode::
 

docs/source/schedulers/lsf.rst

@@ -0,0 +1,18 @@
+IBM Spectrum LSF
+=================
+
+.. automodule:: torchx.schedulers.lsf_scheduler
+
+.. currentmodule:: torchx.schedulers.lsf_scheduler
+
+.. autoclass:: LsfScheduler
+   :members:
+   :show-inheritance:
+
+.. autoclass:: LsfBsub
+   :members:
+
+Reference
+~~~~~~~~~~~~
+
+.. autofunction:: create_scheduler

scripts/component_integration_tests.py

@@ -51,7 +51,7 @@ def main() -> None:
     torchx_image = "dummy_image"
     dryrun = False
 
-    if scheduler in ("kubernetes", "local_docker", "aws_batch"):
+    if scheduler in ("kubernetes", "local_docker", "aws_batch", "lsf"):
         try:
             build = build_and_push_image()
             torchx_image = build.torchx_image
@@ -105,6 +105,17 @@ def main() -> None:
             },
             "workspace": f"file://{os.getcwd()}",
         },
+        "lsf": {
+            "providers": [
+                component_provider,
+            ],
+            "image": torchx_image,
+            "cfg": {
+                "runtime": "docker",
+                "jobdir": "/mnt/data/torchx",
+                "host_network": True,
+            },
+        },
     }
 
     params = run_parameters[scheduler]

torchx/schedulers/__init__.py

@@ -19,6 +19,7 @@
     "kubernetes": "torchx.schedulers.kubernetes_scheduler",
     "aws_batch": "torchx.schedulers.aws_batch_scheduler",
     "ray": "torchx.schedulers.ray_scheduler",
+    "lsf": "torchx.schedulers.lsf_scheduler",
 }