Papers › Dynamic Mini-batch SGD for Elastic Distributed Training: Learning in the Limbo of Resources

Dynamic Mini-batch SGD for Elastic Distributed Training: Learning in the Limbo of Resources

26 Apr 2019arXiv:1904.12043archive 2025-07-28

Haibin Lin, Hang Zhang, Yifei Ma, Tong He, Zhi Zhang, Sheng Zha, Mu Li

With an increasing demand for training powers for deep learning algorithms and the rapid growth of computation resources in data centers, it is desirable to dynamically schedule different distributed deep learning tasks to maximize resource utilization and reduce cost. In this process, different tasks may receive varying numbers of machines at different time, a setting we call elastic distributed training. Despite the recent successes in large mini-batch distributed training, these methods are rarely tested in elastic distributed training environments and suffer degraded performance in our experiments, when we adjust the learning rate linearly immediately with respect to the batch size. One difficulty we observe is that the noise in the stochastic momentum estimation is accumulated over time and will have delayed effects when the batch size changes. We therefore propose to smoothly adjust the learning rate over time to alleviate the influence of the noisy momentum estimation. Our experiments on image classification, object detection and semantic segmentation have demonstrated that our proposed Dynamic SGD method achieves stabilized performance when varying the number of GPUs from 8 to 128. We also provide theoretical understanding on the optimality of linear learning rate scheduling and the effects of stochastic momentum.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 20 code samples harvested from 2 repositories linked to this paper: 13 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 0 violated, 13 with no contract checked; 0 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: community (archive-listed): 20 samples from 2 repositories, 13 ran (13 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-24 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

awslabs/dynamic-training-with-apache-mxnet-on-aws mentioned on GitHubmxnetApache-2.0Syntology: 14 samples, 9 ran (0 instrument-failed) · 5 unverified report
AlanFokCo/compensation-tools tfApache-2.0Syntology: 6 samples, 4 ran (0 instrument-failed) · 2 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

20 samples harvested: 13 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 0 violated, 13 with no contract checked; 0 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

13ran
7unverified

Licence: 14 of the 20 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

awslabs/dynamic-training-with-apache-mxnet-on-aws community (archive-listed) 14 samples, 9 ran (0 instrument-failed) · 5 unverified
bulk python/mxnet/engine.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · f235e1530ce3d53d · report fixture: signature size:int/0/int · returned _BulkScope · level recorded 2026-09-24
c_array in 8 papers amalgamation/python/mxnet_predict.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · 3bc3ed8fa916a182 · report fixture: signature ctype:callable/0/?, values:list/1/int · returned c_int_Array_5 · level recorded 2026-09-25
c_array in 8 papers python/mxnet/base.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · c037ab026fd36b3a · report fixture: signature ctype:callable/0/?, values:list/1/float · returned c_float_Array_4 · level recorded 2026-09-25
c_str in 5 papers amalgamation/python/mxnet_predict.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · 313d72408f5f6086 · report fixture: signature string:str/0/str · returned c_char_p · level recorded 2026-09-25
classproperty python/mxnet/base.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · 47ebd1622534322e · report fixture: signature func:classmethod/0/? · returned _MXClassPropertyDescriptor · level recorded 2026-09-24
do_checkpoint python/mxnet/callback.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · df77b3eca6bd676b · report fixture: signature prefix:str/0/str, period:int/0/int · returned function · level recorded 2026-09-25
log_train_metric python/mxnet/callback.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · 5bc9dd9627f0eb1c · report fixture: signature period:int/0/int, auto_reset:bool/0/bool · returned function · level recorded 2026-09-25
module_checkpoint python/mxnet/callback.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · 4950353b99f3413d · report fixture: signature mod:callable/0/?, prefix:str/0/str, period:int/0/int, save_optimizer_states:bool/0/bool · returned function · level recorded 2026-09-25
record in 8 papers python/mxnet/autograd.py ran Apache-2.0 recorded; this copy not marked cleared · pointer only · b86308bc3bc53b49 · report fixture: signature train_mode:bool/0/bool · returned _RecordingStateScope · level recorded 2026-09-25
create in 17 papers python/mxnet/kvstore.py unverified Apache-2.0 recorded; this copy not marked cleared · pointer only · 9ff2673d6a779a1a · report
set_bulk_size python/mxnet/engine.py unverified Apache-2.0 recorded; this copy not marked cleared · pointer only · 8bcbec3a30a50379 · report
set_recording python/mxnet/autograd.py unverified Apache-2.0 recorded; this copy not marked cleared · pointer only · 84a2b5addc51f455 · report
set_training python/mxnet/autograd.py unverified Apache-2.0 recorded; this copy not marked cleared · pointer only · 565479a708f0d1c6 · report
with_metaclass python/mxnet/base.py unverified Apache-2.0 recorded; this copy not marked cleared · pointer only · 05550ef10b65470b · report
AlanFokCo/compensation-tools community (archive-listed) 6 samples, 4 ran (0 instrument-failed) · 2 unverified
load_data in 700 papers elastic_demos/data_process.py ran Apache-2.0 (permissive) · 3c9ed3332c1f6ca8 · report fixture: signature path:str/0/str · returned tuple · level recorded 2026-09-24
load_data_single elastic_demos/add_data.py ran Apache-2.0 (permissive) · 2d6b2877daa0effe · report fixture: signature file:str/0/str · returned tuple · level recorded 2026-09-24
map_func elastic_demos/tensorflow_v2_cifar_resnet.py ran Apache-2.0 (permissive) · d46ae594e2adad67 · report fixture: signature example:bytes/0/? · returned tuple · level recorded 2026-09-24
unpickle in 45 papers elastic_demos/add_data.py ran Apache-2.0 (permissive) · 91b57896044b9479 · report fixture: signature file:str/0/str · returned dict · level recorded 2026-09-24
load_val_data elastic_demos/data_process.py unverified Apache-2.0 (permissive) · e987ba7540887ac6 · report

Tasks

Image ClassificationObject DetectionSchedulingSemantic Segmentationimage-classificationobject-detection

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

SGD

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections