Skip to content

[WIP] Update benchmark to include chunking options #383

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
bennahugo wants to merge 2 commits into
base: master
Choose a base branch
from
Open

[WIP] Update benchmark to include chunking options #383

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0875232
Update benchmark to include chunking options
bennahugo Mar 7, 2025
08eff6c
Remove debug tracer import
bennahugo Mar 7, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
28 changes: 24 additions & 4 deletions scripts/mvf_read_benchmark.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,6 +34,9 @@
parser.add_argument('--joint', action='store_true', help='Load vis, weights, flags together')
parser.add_argument('--applycal', help='Calibration solutions to apply')
parser.add_argument('--workers', type=int, help='Number of dask workers')
parser.add_argument('--blperchunk', type=int, help='Adjust chunks to specified number of baselines per chunk')
parser.add_argument('--chperchunk', type=int, help='Adjust chunks to specified number of channels per chunk')

args = parser.parse_args()

logging.basicConfig(level='INFO', format='%(asctime)s [%(levelname)s] %(message)s')
Expand All @@ -51,7 +54,24 @@
f.select(dumps=np.s_[:args.dumps])
# Trigger creation of the dask graphs, population of sensor cache for applycal etc
_ = (f.vis[0, 0, 0], f.weights[0, 0, 0], f.flags[0, 0, 0])
nblc = min(args.blperchunk if args.blperchunk else f.vis.dataset.chunksize[0],
f.vis.dataset.shape[0])
nch = min(args.chperchunk if args.chperchunk else f.vis.dataset.chunksize[1],
f.vis.dataset.shape[1])
cs = f.vis.dataset.chunksize
cs = tuple([nblc, nch, cs[2]])
f.vis.dataset.rechunk(cs)
f.weights.dataset.rechunk(cs)
f.flags.dataset.rechunk(cs)
csMB = np.prod(tuple([*cs[0:2], args.time])) * (f.vis.dataset.nbytes // f.vis.dataset.size) / 1024.0**2
visshpGB = f.vis.dataset.nbytes / 1024.0**3
logging.info(f'Selected visibility chunk size {csMB:.2f} MiB of '
f'total selection size {visshpGB:.2f} GiB')
logging.info('Selection complete')
chunk_sizeB = f.vis.dataset.nbytes // f.vis.dataset.size + \
f.weights.dataset.nbytes // f.weights.dataset.size + \
f.flags.dataset.nbytes // f.weights.dataset.size

start = time.time()
last_time = start
for st in range(0, f.shape[0], args.time):
Expand All @@ -65,8 +85,8 @@
current_time = time.time()
elapsed = current_time - last_time
last_time = current_time
size = np.prod(vis.shape) * 10
logging.info('Loaded %d dumps (%.3f MB/s)', vis.shape[0], size / elapsed / 1e6)
size = np.prod(f.shape) * 10
size = np.prod(vis.shape) * chunk_sizeB / 1024.**2
logging.info('Loaded %d dumps (%.3f MiB/s)', vis.shape[0], size / elapsed)
size = np.prod(f.shape) * chunk_sizeB / 1024.**2
elapsed = time.time() - start
logging.info('Loaded %d bytes in %.3f s (%.3f MB/s)', size, elapsed, size / elapsed / 1e6)
logging.info('Loaded %d bytes in %.3f s (%.3f MiB/s)', size, elapsed, size / elapsed)