OpenBox: A Generalized Black-box Optimization Service

Li, Yang; Shen, Yu; Zhang, Wentao; Chen, Yuanwei; Jiang, Huaijun; Liu, Mingchao; Jiang, Jiawei; Gao, Jinyang; Wu, Wentao; Yang, Zhi; Zhang, Ce; Cui, Bin

doi:10.1145/3447548.3467061

Cited by 48 publications

(20 citation statements)

References 30 publications

(35 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DDPG is implemented using PyTorch library [64] with its neural network architecture borrowed from CDBTune [91]. We adopt the OpenBox [50]'s implementation for mixed-kernel BO and GA. For SMAC [34] and TurBO [22], we adopt the implementation released by the authors. Knowledge transfer frameworks.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Facilitating Database Tuning with Hyper-Parameter Optimization: A Comprehensive Experimental Evaluation

Zhang¹,

Chang²,

Yang³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Recently, using automatic configuration tuning to improve the performance of modern database management systems (DBMSs) has attracted increasing interest from the database community. This is embodied with a number of systems featuring advanced tuning capabilities being developed. However, it remains a challenge to select the best solution for database configuration tuning, considering the large body of algorithm choices. In addition, beyond the applications on database systems, we could find more potential algorithms designed for configuration tuning. To this end, this paper provides a comprehensive evaluation of configuration tuning techniques from a broader perspective, hoping to better benefit the database community. In particular, we summarize three key modules of database configuration tuning systems and conduct extensive ablation studies using various challenging cases. Our evaluation demonstrates that the hyper-parameter optimization algorithms can be borrowed to further enhance the database configuration tuning. Moreover, we identify the best algorithm choices for different modules. Beyond the comprehensive evaluations, we offer an efficient and unified database configuration tuning benchmark via surrogates that reduces the evaluation cost to a minimum, allowing for extensive runs and analysis of new techniques.

show abstract

Section: Methodsmentioning

confidence: 99%

“…For workload mapping, we implement the methodology in OtterTune. For RGPE [26], we adopt the implementation in OpenBox [50]. We fine-tune DDPG's pretrained models' weight when obtaining the target observations.…”

Section: Methodsmentioning

confidence: 99%

Facilitating Database Tuning with Hyper-Parameter Optimization: A Comprehensive Experimental Evaluation

Zhang¹,

Chang²,

Yang³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…For MovieLens, ZOOMER and all graph-based baselines use one-hop neighbors for aggregation. We use OpenBox [45] or follow the original papers to get the optimal hyperparameters of our baselines. For MCCF and FGNN, regulation loss weight is set to 5e-7.…”

Section: A Experimental Setupmentioning

confidence: 99%

ZOOMER: Boosting Retrieval on Web-scale Graphs by Regions of Interest

Jiang¹,

Yu²,

Zhao³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

We introduce ZOOMER, a system deployed at Taobao, the largest e-commerce platform in China, for training and serving GNN-based recommendations over web-scale graphs. ZOOMER is designed for tackling two challenges presented by the massive user data at Taobao: low training/serving efficiency due to the huge scale of the graphs, and low recommendation quality due to the information overload which distracts the recommendation model from specific user intentions. ZOOMER achieves this by introducing a key concept, Region of Interests (ROI) in GNNs for recommendations, i.e., a neighborhood region in the graph with significant relevance to a strong user intention. ZOOMER narrows the focus from the whole graph and "zooms in" on the more relevant ROIs, thereby reducing the training/serving cost and mitigating the information overload at the same time. With carefully designed mechanisms, ZOOMER identifies the interest expressed by each recommendation request, constructs an ROI subgraph by sampling with respect to the interest, and guides the GNN to reweigh different parts of the ROI towards the interest by a multi-level attention module. Deployed as a largescale distributed system, ZOOMER supports graphs with billions of nodes for training and thousands of requests per second for serving. ZOOMER achieves up to 14x speedup when downsizing sampling scales with comparable (even better) AUC performance than baseline methods. Besides, both the offline evaluation and online A/B test demonstrate the effectiveness of ZOOMER.

show abstract

“…The experiments are implemented on 4 machines with 14 Intel(R) Xeon(R) CPUs (Gold 5120 @ 2.20GHz) and four NVIDIA TITAN RTX GPUs. The code is written in Python 3.6, and the multi-objective algorithm is implemented based on OpenBox [25]. We use Pytorch 1.7.1 on CUDA 10.1 to train the model on GPU.…”

Section: A4 Experiments Setupmentioning

confidence: 99%

PaSca: a Graph Neural Architecture Search System under the Scalable Paradigm

Zhang,

Shen,

Lin

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Graph neural networks (GNNs) have achieved state-of-the-art performance in various graph-based tasks. However, as mainstream GNNs are designed based on the neural message passing mechanism, they do not scale well to data size and message passing steps. Although there has been an emerging interest in the design of scalable GNNs, current researches focus on specific GNN design, rather than the general design space, limiting the discovery of potential scalable GNN models. This paper proposes PaSca, a new paradigm and system that offers a principled approach to systemically construct and explore the design space for scalable GNNs, rather than studying individual designs. Through deconstructing the message passing mechanism, PaSca presents a novel Scalable Graph Neural Architecture Paradigm (SGAP), together with a general architecture design space consisting of 150k different designs. Following the paradigm, we implement an auto-search engine that can automatically search well-performing and scalable GNN architectures to balance the trade-off between multiple criteria (e.g., accuracy and efficiency) via multi-objective optimization. Empirical studies on ten benchmark datasets demonstrate that the representative instances (i.e., PaSca-V1, V2, and V3) discovered by our system achieve consistent performance among competitive baselines. Concretely, PaSca-V3 outperforms the state-of-the-art GNN method JK-Net by 0.4% in terms of predictive accuracy on our large industry dataset while achieving up to 28.3× training speedups. CCS CONCEPTS• Mathematics of computing → Graph algorithms.

show abstract

OpenBox: A Generalized Black-box Optimization Service

Cited by 48 publications

References 30 publications

Facilitating Database Tuning with Hyper-Parameter Optimization: A Comprehensive Experimental Evaluation

Facilitating Database Tuning with Hyper-Parameter Optimization: A Comprehensive Experimental Evaluation

ZOOMER: Boosting Retrieval on Web-scale Graphs by Regions of Interest

PaSca: a Graph Neural Architecture Search System under the Scalable Paradigm

Contact Info

Product

Resources

About