ray/python at af0c1174cd35a65ec3ca3630b999577fd4614578 - ray

mirror of https://github.com/wassname/ray.git synced 2026-06-28 04:07:01 +08:00

Files

T

Eric Liang af0c1174cd [sgd] Merge sharded param server based SGD implementation (#3033 )

This includes most of the TF code used for the OSDI experiment. Perf sanity check on p3.16xl instances: Overall scaling looks ok, with the multi-node results within 5% of OSDI final numbers. This seems reasonable given that hugepages are not enabled here, and the param server shards are placed randomly.

$ RAY_USE_XRAY=1 ./test_sgd.py --gpu --batch-size=64 --num-workers=N \
  --devices-per-worker=M --strategy=<simple|ps> \
  --warmup --object-store-memory=10000000000

Images per second total
gpus total              | simple | ps
========================================
1                       | 218
2 (1 worker)            | 388
4 (1 worker)            | 759
4 (2 workers)           | 176    | 623
8 (1 worker)            | 985
8 (2 workers)           | 349    | 1031
16 (2 nodes, 2 workers) | 600    | 1661
16 (2 nodes, 4 workers) | 468    | 1712   <--- OSDI perf was 1817

2018-10-27 21:25:02 -07:00

benchmarks

[asv] Add benchmark for ray.wait (#2625 )

2018-08-10 17:52:36 -07:00

ray

[sgd] Merge sharded param server based SGD implementation (#3033 )

2018-10-27 21:25:02 -07:00

asv.conf.json

[asv] Pushing to s3 (#2246 )

2018-06-20 10:43:44 -07:00

build-wheel-macos.sh

Adding Python3.7 wheels support (#2546 )