SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.02023 · NeurIPS · 2022

DreamShard: Generalizable Embedding Table Placement for Recommender Systems

Daochen Zha, Xia Hu, Yuandong Tian, Louis Feng, Qiaoyu Tan, Zirui Liu, Kwei-Herng Lai, Bhargav Bhushanam, Arun Kejariwal

arXiv · PDF · Open in the Atlas

Code that ran

2 of 4 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 2 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2210.02023.json · badge: /api/ran/2210.02023.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
daochenzha/dreamshard canonical (classifier label)
classifier basis: classifier_2025
2 ran · 2 no recorded run
FunctionStatusWhere it lives
plan2allocation Ran
ran, honoured the contract we drafted
daochenzha/dreamshard/dreamshard/sharders.py
code served (permissive licence) · get_code("3146017b0709fbef")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Analyzing plan2allocation:
# - Takes 'plan' as parameter
# - plan is iterated: sum([len(shard) for shard in plan])
# - plan contains partitions (shards)
# - Each partition is iterable and contains indices
# - table_device_indices is indexed with these indices
# - bin_id is used as device assignment

# Construct a valid plan
# plan is a list of partitions, where each partition is a list of table indices
plan = [
    [0, 1],      # shard 0: tables 0 and 1
    [2, 3, 4],   # shard 1: tables 2, 3, 4
    [5]          # shard 2: table 5
]

ARGS = {
    'plan': plan
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family int — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
register_sharder Ran
ran; our contract draft was wrong, not the code
daochenzha/dreamshard/dreamshard/sharders.py
code served (permissive licence) · get_code("1853184b9cb1b65f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned function. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The register_sharder function is a decorator factory
# It takes a sharder_name (string) and returns a decorator
# The decorator wraps a function and registers it in _sharders dict
# We need to provide a sharder_name string

sharder_name = "test_sharder"

ARGS = {
    "sharder_name": sharder_name
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
dreamshard Not yet run
unverified (harvested, no recorded run)
daochenzha/dreamshard/dreamshard/sharders.py
code served (permissive licence) · get_code("fb67e93fab6a305e")
How it ran

no invocation recorded.

shard Not yet run
unverified (harvested, no recorded run)
daochenzha/dreamshard/dreamshard/sharders.py
code served (permissive licence) · get_code("12b7f16889cfe1f2")
How it ran

no invocation recorded.

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 3 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
register_sharder
brick of this paper's assembly: helper
1 other paper
1 later
2303.10158 (2023-03)
Data-centric Artificial Intelligence: A Survey

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We study embedding table placement for distributed recommender systems, which aims to partition and place the tables on multiple hardware devices (e.g., GPUs) to balance the computation and communication costs. Although prior work has explored learning-based approaches for the device placement of computational graphs, embedding table placement remains to be a challenging problem because of 1) the operation fusion of embedding tables, and 2) the generalizability requirement on unseen placement tasks with different numbers of tables and/or devices. To this end, we present DreamShard, a reinforcement learning (RL) approach for embedding table placement. DreamShard achieves the reasoning of operation fusion and generalizability with 1) a cost network to directly predict the costs of the fused operation, and 2) a policy network that is efficiently trained on an estimated Markov decision process (MDP) without real GPU execution, where the states and the rewards are estimated with the cost network. Equipped with sum and max representation reductions, the two networks can directly generalize to any unseen tasks with different numbers of tables and/or devices without finetuning. Extensive experiments show that DreamShard substantially outperforms the existing human expert and RNN-based strategies with up to 19% speedup over the strongest baseline on large-scale synthetic tables and our production tables. The code is available at https://github.com/daochenzha/dreamshard.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2210.02023")
get_code_for_paper("2210.02023")
have("2210.02023")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2210.02023.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2210.02023.svg)](https://syntology.ai/paper/2210.02023)

Connect an agent — have() is free.