Writing large number of metrics to Google Cloud Storage via Cloud Storage FUSE is very slow
I maintainer di solito rispondono entro 1 giorno
Nessuno ha ancora preso questa issue.
Valutazione
- Difficoltà
- 4/5
- Tempo stimato
- 3-5 giorni
- Idoneità per principianti
- 38/100
- Tipo di issue
- Bug
- Chiarezza
- Abbastanza chiara
- Stato di attività
- Ferma
- Stack tecnologico
- google-cloud, python
- Ambito
- cloud, performance
Direzione di ricerca
Inizia da tensorboard.summary.writer.event_file_writer.EventFileWriter e verifica come il suo writer di file generico gestisce fs_supports_append. Riproduci il confronto con un percorso Google Cloud Storage FUSE, quindi verifica che la scrittura di molti eventi eviti il ritardo dimostrato preservando il normale comportamento del filesystem locale; il payload non nomina alcun file di test.
Scritto dal modello di indicizzazione a partire dal testo della issue.
Descrizione
This can be trivially reproduced given that you have an access to some Google Cloud project in which you have permissions to run Vertex AI custom training jobs and create Google Cloud Storage (GCS) buckets.
- In Vertex AI, submit some fake custom training job (using e.g.
python:3.10Docker image andcommand: ["sleep", "3600"]) with interactive shell enabled (if you're submitting this job via Python's SDK, simply passenable_web_access=TruetoCustomJob's.submit()method). - Once you'd get this job in "Training" state, open two terminals by clicking "Launch web terminal" in its GUI - we will use them for Python and Bash, respectively.
- In GCS, create some bucket of your choice and make sure that you can write to it.
- In the first terminal, run
pip install tensorboardand then launch Python interpreter. - Copy & paste this code to Python's shell, replacing
<NAME_OF_YOUR_BUCKET>with the name of the bucket that you've created in step (3):
from time import time
from tensorboard.compat.proto import event_pb2
from tensorboard.compat.proto.summary_pb2 import Summary
from tensorboard.summary.writer.event_file_writer import EventFileWriter
path = "/gcs/<NAME_OF_YOUR_BUCKET>/"
def get_fake_metric(i):
summary = Summary(value=[Summary.Value(tag=f"val_metric{i}", simple_value=0.123)])
event = event_pb2.Event(summary=summary)
event.wall_time = time()
event.step = 1
return event
- In the second terminal (Bash),
cdto/gcs/<NAME_OF_YOUR_BUCKET>/and make sure that there are no existing event files there. - We will start with the default writer - paste this to Python's shell:
writer = EventFileWriter(path)
- Switch to Bash and run the following (we will be using this to monitor the size of event files - at the beginning you should see one such file having 88 bytes):
while true; do date '+%H:%M:%S'; ls -l | grep event; sleep 1; done
- Switch to Python and run:
for i in range(200): writer.add_event(get_fake_metric(i))
- Switch to Bash and observe how your event file is growing - it should reach slightly above 10 KB. Once it will stop growing, terminate your monitoring loop and note how long it took (for me, it was
2m23s). Delete this file afterwards. - Now, we will repeat this experiment with slightly modified writer. Switch to Python and paste the following:
writer = EventFileWriter(path)
assert writer._general_file_writer.fs_supports_append
writer._general_file_writer.fs_supports_append = False
- Switch to Bash and run the same monitoring loop as in step (8).
- Switch to Python and run the same loop as in step (9), followed by
writer.flush()because otherwise you'd have to wait extra 2 minutes. - Switch to Bash and notice that your event file reached its expected size almost instantly.
Q: What does this experiment show?
A: It shows that EventFileWriter is not aware of Cloud Storage FUSE filesystem, which leads to suboptimal performance (and to make it clear: this monkey-patching of fs_supports_append doesn't mean that this fs doesn't support append - it's just a hack meant to foolEventFileWriter).
Q: But writing events to disk is asynchronous, so even if it is slow, why should we care?
A: Because some training frameworks (e.g. PyTorch Lightning) explicitly flush metrics at the end of the validation loop (which is quite reasonable thing to do, I'd say), and with hundreds of them (600+ in my case), training is stuck for non-negligible time after each validation (7-8 minutes in my case) - and the more frequent the validations, the more badly it accumulates w.r.t. total training time and GPU utilization. And besides, good engineers simply can't stand seeing that writing such relatively small amount of data takes so much time ;)
- Lingua principale
- TypeScript
- Stelle
- 7.2k
- Fork
- 1.7k
- Merge medio
- 1g 13h
- PR unite (30g)
- 3
Preparare l'ambiente
- Include un Dockerfile o un file Docker Compose
- Ha un modello di pull request
- Leggi la guida per i contributori
Come iniziare
- Leggi tutta la issue e poi la guida ai contributi del progetto.
- Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
- Fai un fork del repository e lavora su un branch.
- Apri una pull request che faccia riferimento al numero della issue.
Altre issue di tensorflow/tensorboard
-
node_repositories in WORKSPACE missing linux_ppc64le and linux_s390x entries for Node.js 18.20.8Aperta
Difficoltà 1/5 Meno di un'ora Idoneità per principianti 92/100
tensorflow/tensorboard#7140 · 1 reazione ·
I maintainer di solito rispondono entro 1 giorno
-
Error Parsing Hyper Parameters (`HPARAMS`) on WindowsForse già presa @Ramnath0521 l’ha presa 4 giorni fa. Aperta
Difficoltà 2/5 1-3 ore Idoneità per principianti 35/100
tensorflow/tensorboard#7175 ·
I maintainer di solito rispondono entro 1 giorno
-
Difficoltà 5/5 Più di una settimana Idoneità per principianti 10/100
tensorflow/tensorboard#7159 ·
I maintainer di solito rispondono entro 1 giorno
-
installAperta
Difficoltà 5/5 Più di una settimana Idoneità per principianti 10/100
tensorflow/tensorboard#7158 ·
I maintainer di solito rispondono entro 1 giorno
-
Difficoltà 4/5 3-5 giorni Idoneità per principianti 35/100
tensorflow/tensorboard#7153 ·
I maintainer di solito rispondono entro 1 giorno
Tutte le issue di tensorflow/tensorboard
Issue simili
-
documentation lens:human needs-triage
Difficoltà 2/5 1-3 ore Idoneità per principianti 73/100
thebristolsound/birdbrain#2246 ·
I maintainer di solito rispondono entro 1 giorno
-
Difficoltà 1/5 Meno di un'ora Idoneità per principianti 78/100
I maintainer di solito rispondono entro 1 giorno
-
A voting window past about 69.9 million hours stores a deadline that reads back as no deadlineApertabug ready-for-human
Difficoltà 2/5 1-3 ore Idoneità per principianti 72/100
BoardGamesEmpire/board-games-empire-backend#1452 · 1 commento ·
I maintainer di solito rispondono entro 1 giorno
-
good first issue stretch
Difficoltà 2/5 1-3 ore Idoneità per principianti 65/100
-
Difficoltà 2/5 1-3 ore Idoneità per principianti 76/100
I maintainer di solito rispondono entro 1 giorno