{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T22:38:00Z","timestamp":1778279880345,"version":"3.51.4"},"reference-count":84,"publisher":"Springer Science and Business Media LLC","issue":"2","license":[{"start":{"date-parts":[[2025,1,31]],"date-time":"2025-01-31T00:00:00Z","timestamp":1738281600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,1,31]],"date-time":"2025-01-31T00:00:00Z","timestamp":1738281600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100005950","name":"Hong Kong University of Science and Technology","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100005950","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["The VLDB Journal"],"published-print":{"date-parts":[[2025,3]]},"abstract":"<jats:title>Abstract<\/jats:title>\n          <jats:p>Graph neural networks (GNNs) have emerged due to their success at modeling graph data. Yet, it is challenging for GNNs to efficiently scale to large graphs. Thus, distributed GNNs come into play. To avoid communication caused by expensive data movement between workers, we propose S<jats:sc>ancus<\/jats:sc> and its advanced version S<jats:sc>ancus<\/jats:sc>\"Equation missing\", the staleness and quantization-aware communication-avoiding decentralized GNN system. By introducing a set of novel bounded embedding staleness metrics and adaptively skipping broadcasts, S<jats:sc>ancus<\/jats:sc> abstracts decentralized GNN processing as sequential matrix multiplication and uses historical embeddings via cache. To further mitigate the communication volume, S<jats:sc>ancus<\/jats:sc>\"Equation missing\" conducts quantization-aware communication on embeddings to reduce the size of broadcast messages. Theoretically, we show bounded approximation errors of embeddings and gradients with a known fastest convergence guarantee. Empirically, we evaluate S<jats:sc>ancus<\/jats:sc> and S<jats:sc>ancus<\/jats:sc>\"Equation missing\" with common GNN models via different system setups on large-scale benchmark datasets. Compared to SOTA works, S<jats:sc>ancus<\/jats:sc>\"Equation missing\" can avoid up to <jats:inline-formula>\n              <jats:alternatives>\n                <jats:tex-math>$$86\\%$$<\/jats:tex-math>\n                <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                  <mml:mrow>\n                    <mml:mn>86<\/mml:mn>\n                    <mml:mo>%<\/mml:mo>\n                  <\/mml:mrow>\n                <\/mml:math>\n              <\/jats:alternatives>\n            <\/jats:inline-formula> communication with <jats:inline-formula>\n              <jats:alternatives>\n                <jats:tex-math>$$3.0\\times $$<\/jats:tex-math>\n                <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                  <mml:mrow>\n                    <mml:mn>3.0<\/mml:mn>\n                    <mml:mo>\u00d7<\/mml:mo>\n                  <\/mml:mrow>\n                <\/mml:math>\n              <\/jats:alternatives>\n            <\/jats:inline-formula> faster throughput on average without accuracy loss.<\/jats:p>","DOI":"10.1007\/s00778-024-00897-2","type":"journal-article","created":{"date-parts":[[2025,1,31]],"date-time":"2025-01-31T21:49:37Z","timestamp":1738360177000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["From Sancus to Sancus $$^q$$: staleness and quantization-aware full-graph decentralized training in graph neural networks"],"prefix":"10.1007","volume":"34","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4121-6284","authenticated-orcid":false,"given":"Jingshu","family":"Peng","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Qiyu","family":"Liu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhao","family":"Chen","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yingxia","family":"Shao","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yanyan","family":"Shen","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lei","family":"Chen","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiannong","family":"Cao","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2025,1,31]]},"reference":[{"issue":"9","key":"897_CR1","doi-asserted-by":"publisher","first-page":"191:1","DOI":"10.1145\/3477141","volume":"54","author":"S Abadal","year":"2022","unstructured":"Abadal, S., Jain, A., Guirado, R., L\u00f3pez-Alonso, J., Alarc\u00f3n, E.: Computing graph neural networks: A survey from algorithms to accelerators. ACM Comput. Surv. 54(9), 191:1-191:38 (2022)","journal-title":"ACM Comput. Surv."},{"key":"897_CR2","doi-asserted-by":"crossref","unstructured":"Aguerrebere, C., Bhati, I.S., Hildebrand, M., Tepper, M., Willke, T.L.: Similarity search in the blink of an eye with compressed indices. Proc. VLDB Endow. 16(11) (2023)","DOI":"10.14778\/3611479.3611537"},{"key":"897_CR3","unstructured":"Alistarh, D., Grubic, D., Li, J., Tomioka, R., Vojnovic, M.: QSGD: communication-efficient SGD via gradient quantization and encoding. In: NIPS (2017)"},{"issue":"4","key":"897_CR4","doi-asserted-by":"publisher","first-page":"288","DOI":"10.14778\/2856318.2856324","volume":"9","author":"F Andr\u00e9","year":"2015","unstructured":"Andr\u00e9, F., Kermarrec, A., Scouarnec, N.L.: Cache locality is not enough: High-performance nearest neighbor search with product quantization fast scan. Proc. VLDB Endow. 9(4), 288\u2013299 (2015)","journal-title":"Proc. VLDB Endow."},{"key":"897_CR5","unstructured":"Bajaj, S., Son, H., Liu, J., Guan, H., Serafini, M.: Graph neural network training systems: A performance comparison of full-graph and mini-batch. arXiv preprint arXiv:2406.00552 (2024)"},{"key":"897_CR6","doi-asserted-by":"crossref","unstructured":"Berchtold, S., B\u00f6hm, C., Jagadish, H.V., Kriegel, H., Sander, J.: Independent quantization: An index compression technique for high-dimensional data spaces. In: ICDE, pp. 577\u2013588. IEEE Computer Society (2000)","DOI":"10.1109\/ICDE.2000.839456"},{"key":"897_CR7","doi-asserted-by":"crossref","unstructured":"Blalock, D.W., Guttag, J.V.: Bolt: Accelerated data mining with fast vector compression. In: KDD. ACM (2017)","DOI":"10.1145\/3097983.3098195"},{"key":"897_CR8","doi-asserted-by":"crossref","unstructured":"Cai, Z., Yan, X., Wu, Y., Ma, K., Cheng, J., Yu, F.: DGCL: an efficient communication library for distributed GNN training. In: EuroSys \u201921: Sixteenth European Conference on Computer Systems, Online Event, United Kingdom, April 26-28, 2021, pp. 130\u2013144. ACM (2021)","DOI":"10.1145\/3447786.3456233"},{"key":"897_CR9","unstructured":"Chakrabarti, A., Moseley, B.: Backprop with approximate activations for memory-efficient network training. In: NeurIPS, pp. 2426\u20132435 (2019)"},{"key":"897_CR10","doi-asserted-by":"crossref","unstructured":"Chan, E., Heimlich, M., Purkayastha, A., van\u00a0de Geijn, R.A.: Collective communication: theory, practice, and experience. Concurr. Comput. Pract. Exp. 19(13) (2007)","DOI":"10.1002\/cpe.1206"},{"key":"897_CR11","unstructured":"Chen, J., Zheng, L., Yao, Z., Wang, D., Stoica, I., Mahoney, M.W., Gonzalez, J.: Actnn: Reducing training memory footprint via 2-bit activation compressed training. In: ICML, vol. 139, pp. 1803\u20131813. PMLR (2021)"},{"key":"897_CR12","unstructured":"Chen, J., Zhu, J., Song, L.: Stochastic training of graph convolutional networks with variance reduction. In: ICML, Proceedings of Machine Learning Research, vol.\u00a080, pp. 941\u2013949. PMLR (2018)"},{"key":"897_CR13","doi-asserted-by":"crossref","unstructured":"Chen, R., Shi, J., Chen, Y., Zang, B., Guan, H., Chen, H.: Powerlyra: Differentiated graph computation and partitioning on skewed graphs. ACM Trans. Parallel Comput. 5(3) (2018)","DOI":"10.1145\/3298989"},{"key":"897_CR14","doi-asserted-by":"crossref","unstructured":"Chen, S., Zheng, D., Ding, C., Huan, C., Ji, Y., Liu, H.: TANGO: re-thinking quantization for graph neural network training on gpus. In: SC. ACM (2023)","DOI":"10.1145\/3581784.3607037"},{"key":"897_CR15","unstructured":"Cipar, J., Ho, Q., Kim, J.K., Lee, S., Ganger, G.R., Gibson, G., Keeton, K., Xing, E.P.: Solving the straggler problem with bounded staleness. In: HotOS. USENIX Association (2013)"},{"key":"897_CR16","doi-asserted-by":"crossref","unstructured":"Cong, W., Forsati, R., Kandemir, M.T., Mahdavi, M.: Minimal variance sampling with provable guarantees for fast training of graph neural networks. In: KDD, pp. 1393\u20131403. ACM (2020)","DOI":"10.1145\/3394486.3403192"},{"key":"897_CR17","doi-asserted-by":"crossref","unstructured":"Croci, M., Fasi, M., Higham, N.J., Mary, T., Mikaitis, M.: Stochastic rounding: implementation, error analysis and applications. Royal Society Open Science 9(3), 211,631 (2022)","DOI":"10.1098\/rsos.211631"},{"key":"897_CR18","unstructured":"Dai, W., Zhou, Y., Dong, N., Zhang, H., Xing, E.P.: Toward understanding the impact of staleness in distributed machine learning. In: ICLR (2019)"},{"key":"897_CR19","unstructured":"Faghri, F., Tabrizian, I., Markov, I., Alistarh, D., Roy, D.M., Ramezani-Kebrya, A.: Adaptive gradient quantization for data-parallel SGD. In: NeurIPS (2020)"},{"issue":"12","key":"897_CR20","doi-asserted-by":"publisher","first-page":"1889","DOI":"10.14778\/3137765.3137801","volume":"10","author":"W Fan","year":"2017","unstructured":"Fan, W., Xu, J., Wu, Y., Yu, W., Jiang, J.: GRAPE: parallelizing sequential graph computations. Proc. VLDB Endow. 10(12), 1889\u20131892 (2017)","journal-title":"Proc. VLDB Endow."},{"key":"897_CR21","doi-asserted-by":"crossref","unstructured":"Fan, W., Xu, J., Wu, Y., Yu, W., Jiang, J., Zheng, Z., Zhang, B., Cao, Y., Tian, C.: Parallelizing sequential graph computations. In: SIGMOD 2017, Chicago, IL, USA, May 14-19. ACM (2017)","DOI":"10.1145\/3035918.3035942"},{"key":"897_CR22","unstructured":"Fey, M., Lenssen, J.E., Weichert, F., Leskovec, J.: Gnnautoscale: Scalable and expressive graph neural networks via historical embeddings. In: ICML, Proceedings of Machine Learning Research, vol. 139. PMLR (2021)"},{"key":"897_CR23","first-page":"61","volume":"56","author":"GE Forsythe","year":"1950","unstructured":"Forsythe, G.E.: Round-off errors in numerical integration on automatic machinery-preliminary report. Bull. Amer. Math. Soc 56, 61\u201362 (1950)","journal-title":"Bull. Amer. Math. Soc"},{"key":"897_CR24","unstructured":"Gandhi, S., Iyer, A.P.: P3: distributed deep graph learning at scale. In: OSDI. USENIX Association (2021)"},{"key":"897_CR25","doi-asserted-by":"crossref","unstructured":"Gao, J., Long, C.: Rabitq: Quantizing high-dimensional vectors with a theoretical error bound for approximate nearest neighbor search. SIGMOD \u201924,Proceedings of the ACM on Management of Data 2(3), 1\u201327 (2024)","DOI":"10.1145\/3654970"},{"key":"897_CR26","doi-asserted-by":"crossref","unstructured":"Gholami, A., Azad, A., Jin, P.H., Keutzer, K., Bulu\u00e7, A.: Integrated model, batch, and domain parallelism in training neural networks. In: SPAA. ACM (2018)","DOI":"10.1145\/3210377.3210394"},{"issue":"2","key":"897_CR27","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1109\/MASSP.1984.1162229","volume":"1","author":"R Gray","year":"1984","unstructured":"Gray, R.: Vector quantization. IEEE Assp Magazine 1(2), 4\u201329 (1984)","journal-title":"Vector quantization. IEEE Assp Magazine"},{"key":"897_CR28","unstructured":"Gupta, S., Agrawal, A., Gopalakrishnan, K., Narayanan, P.: Deep learning with limited numerical precision. In: ICML, vol.\u00a037, pp. 1737\u20131746. JMLR.org (2015)"},{"issue":"1","key":"897_CR29","doi-asserted-by":"publisher","first-page":"91","DOI":"10.1080\/10556788.2022.2117355","volume":"38","author":"S Horv\u00e1th","year":"2023","unstructured":"Horv\u00e1th, S., Kovalev, D., Mishchenko, K., Richt\u00e1rik, P., Stich, S.U.: Stochastic distributed learning with gradient quantization and double-variance reduction. Optim. Methods Softw. 38(1), 91\u2013106 (2023)","journal-title":"Optim. Methods Softw."},{"key":"897_CR30","unstructured":"Hu, W., Fey, M., Zitnik, M., Dong, Y., Ren, H., Liu, B., Catasta, M., Leskovec, J.: Open graph benchmark: Datasets for machine learning on graphs. In: NeurIPS (2020)"},{"key":"897_CR31","doi-asserted-by":"crossref","unstructured":"Huskey, H.D., Hartree, D.R.: On the precision of a certain procedure of numerical integration. Department of Commerce (1949)","DOI":"10.6028\/jres.042.005"},{"key":"897_CR32","unstructured":"Jia, Z., Lin, S., Gao, M., Zaharia, M., Aiken, A.: Improving the accuracy, scalability, and performance of graph neural networks with roc. In: MLSys. mlsys.org (2020)"},{"key":"897_CR33","doi-asserted-by":"crossref","unstructured":"Jiang, J., Cui, B., Zhang, C., Yu, L.: Heterogeneity-aware distributed parameter servers. In: SIGMOD Conference, pp. 463\u2013478. ACM (2017)","DOI":"10.1145\/3035918.3035933"},{"issue":"5","key":"897_CR34","doi-asserted-by":"publisher","first-page":"945","DOI":"10.1007\/s00778-019-00596-3","volume":"29","author":"J Jiang","year":"2020","unstructured":"Jiang, J., Fu, F., Yang, T., Shao, Y., Cui, B.: Skcompress: compressing sparse and nonuniform gradient in distributed machine learning. VLDB J. 29(5), 945\u2013972 (2020)","journal-title":"VLDB J."},{"key":"897_CR35","doi-asserted-by":"crossref","unstructured":"Karypis, G., Kumar, V.: A fast and high quality multilevel scheme for partitioning irregular graphs. SIAM J. Sci. Comput. 20(1) (1998)","DOI":"10.1137\/S1064827595287997"},{"key":"897_CR36","doi-asserted-by":"crossref","unstructured":"Kim, W., Lochovsky, F.H. (eds.): Object-Oriented Concepts, Databases, and Applications. ACM Press and Addison-Wesley (1989)","DOI":"10.1145\/63320"},{"key":"897_CR37","unstructured":"Kipf, T.N., Welling, M.: Semi-supervised classification with graph convolutional networks. In: ICLR (Poster). OpenReview.net (2017)"},{"key":"897_CR38","unstructured":"lambda: Lambda (2022). https:\/\/aws.amazon.com\/lambda\/"},{"key":"897_CR39","doi-asserted-by":"crossref","unstructured":"Lee, J.H., Kim, H.: Empirical investigation of stale value tolerance on parallel RNN training. In: ISPASS, pp. 153\u2013164. IEEE (2019)","DOI":"10.1109\/ISPASS.2019.00029"},{"key":"897_CR40","doi-asserted-by":"crossref","unstructured":"Li, H., Kadav, A., Kruus, E., Ungureanu, C.: MALT: distributed data-parallelism for existing ML applications. In: EuroSys, pp. 3:1\u20133:16. ACM (2015)","DOI":"10.1145\/2741948.2741965"},{"key":"897_CR41","volume-title":"Can decentralized algorithms outperform centralized algorithms?","author":"X Lian","year":"2017","unstructured":"Lian, X., Zhang, C., Zhang, H., Hsieh, C., Zhang, W., Liu, J.: Can decentralized algorithms outperform centralized algorithms? NIPS, A case study for decentralized parallel stochastic gradient descent. In (2017)"},{"issue":"3","key":"897_CR42","doi-asserted-by":"publisher","first-page":"667","DOI":"10.1007\/s00778-023-00829-6","volume":"33","author":"N Liao","year":"2024","unstructured":"Liao, N., Mo, D., Luo, S., Li, X., Yin, P.: Scalable decoupling graph neural network with feature-oriented optimization. The VLDB Journal 33(3), 667\u2013683 (2024)","journal-title":"The VLDB Journal"},{"key":"897_CR43","first-page":"87","volume":"6","author":"J Lin","year":"2024","unstructured":"Lin, J., Tang, J., Tang, H., Yang, S., Chen, W.M., Wang, W.C., Xiao, G., Dang, X., Gan, C., Han, S.: Awq: Activation-aware weight quantization for on-device llm compression and acceleration. Proceedings of Machine Learning and Systems 6, 87\u2013100 (2024)","journal-title":"Proceedings of Machine Learning and Systems"},{"issue":"3","key":"897_CR44","doi-asserted-by":"publisher","first-page":"278","DOI":"10.14778\/3632093.3632095","volume":"17","author":"Q Lin","year":"2023","unstructured":"Lin, Q., Wu, S., Zhao, J., Dai, J., Shi, M., Chen, G., Li, F.: Smartlite: A dbms-based serving system for DNN inference in resource-constrained environments. Proc. VLDB Endow. 17(3), 278\u2013291 (2023)","journal-title":"Proc. VLDB Endow."},{"key":"897_CR45","doi-asserted-by":"crossref","unstructured":"Lin, Z., Li, C., Miao, Y., Liu, Y., Xu, Y.: Pagraph: Scaling GNN training on large graphs via computation-aware caching. In: SoCC, pp. 401\u2013415. ACM (2020)","DOI":"10.1145\/3419111.3421281"},{"key":"897_CR46","doi-asserted-by":"crossref","unstructured":"Liu, H., Lu, S., Chen, X., He, B.: G3: when graph neural networks meet parallel graph processing systems on gpus. Proc. VLDB Endow. 13(12) (2020)","DOI":"10.14778\/3415478.3415482"},{"key":"897_CR47","unstructured":"Liu, Z., Zhou, K., Yang, F., Li, L., Chen, R., Hu, X.: EXACT: scalable graph neural networks training via extreme activation compression. In: ICLR (2022)"},{"key":"897_CR48","doi-asserted-by":"crossref","unstructured":"Lu, Y., Kandula, S., K\u00f6nig, A.C., Chaudhuri, S.: Pre-training summarization models of structured datasets for cardinality estimation. Proc. VLDB Endow. 15(3) (2021)","DOI":"10.14778\/3494124.3494127"},{"key":"897_CR49","doi-asserted-by":"crossref","unstructured":"Luo, Q., He, J., Zhuo, Y., Qian, X.: Prague: High-performance heterogeneity-aware asynchronous decentralized training. In: ASPLOS, pp. 401\u2013416. ACM (2020)","DOI":"10.1145\/3373376.3378499"},{"key":"897_CR50","unstructured":"Ma, L., Yang, Z., Miao, Y., Xue, J., Wu, M., Zhou, L., Dai, Y.: Neugraph: Parallel deep neural network computation on large graphs. In: USENIX Annual Technical Conference, pp. 443\u2013458. USENIX Association (2019)"},{"key":"897_CR51","doi-asserted-by":"crossref","unstructured":"Malewicz, G., Austern, M.H., Bik, A.J.C., Dehnert, J.C., Horn, I., Leiser, N., Czajkowski, G.: Pregel: a system for large-scale graph processing. In: SIGMOD Conference, pp. 135\u2013146. ACM (2010)","DOI":"10.1145\/1807167.1807184"},{"key":"897_CR52","doi-asserted-by":"crossref","unstructured":"Miao, X., Nie, X., Shao, Y., Yang, Z., Jiang, J., Ma, L., Cui, B.: Heterogeneity-aware distributed machine learning training via partial reduce. In: SIGMOD Conference, pp. 2262\u20132270. ACM (2021)","DOI":"10.1145\/3448016.3452773"},{"issue":"11","key":"897_CR53","doi-asserted-by":"publisher","first-page":"2087","DOI":"10.14778\/3476249.3476264","volume":"14","author":"S Min","year":"2021","unstructured":"Min, S., Wu, K., Huang, S., Hidayetoglu, M., Xiong, J., Ebrahimi, E., Chen, D., Hwu, W.W.: Large graph convolutional network training with gpu-oriented data communication architecture. Proc. VLDB Endow. 14(11), 2087\u20132100 (2021)","journal-title":"Proc. VLDB Endow."},{"key":"897_CR54","unstructured":"Nvidia, C.: cuRAND library programming guide. https:\/\/developer.nvidia.com\/curand (2024)"},{"key":"897_CR55","doi-asserted-by":"crossref","unstructured":"Paparrizos, J., Edian, I., Liu, C., Elmore, A.J., Franklin, M.J.: Fast adaptive similarity search through variance-aware quantization. In: ICDE. IEEE (2022)","DOI":"10.1109\/ICDE53745.2022.00268"},{"issue":"11","key":"897_CR56","doi-asserted-by":"publisher","first-page":"2626","DOI":"10.14778\/3551793.3551819","volume":"15","author":"Y Park","year":"2022","unstructured":"Park, Y., Min, S., Lee, J.W.: Ginex: Ssd-enabled billion-scale graph neural network training on a single machine via provably optimal in-memory caching. Proc. VLDB Endow. 15(11), 2626\u20132639 (2022)","journal-title":"Proc. VLDB Endow."},{"key":"897_CR57","unstructured":"Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., Desmaison, A., K\u00f6pf, A., Yang, E.Z., DeVito, Z., Raison, M., Tejani, A., Chilamkurthy, S., Steiner, B., Fang, L., Bai, J., Chintala, S.: Pytorch: An imperative style, high-performance deep learning library. In: NeurIPS, pp. 8024\u20138035 (2019)"},{"key":"897_CR58","doi-asserted-by":"crossref","unstructured":"Sha, M., Li, Y., He, B., Tan, K.: Accelerating dynamic graph analytics on gpus. Proc. VLDB Endow. 11(1) (2017)","DOI":"10.14778\/3151113.3151122"},{"key":"897_CR59","unstructured":"Thorpe, J., Qiao, Y., Eyolfson, J., Teng, S., Hu, G., Jia, Z., Wei, J., Vora, K., Netravali, R., Kim, M., Xu, G.H.: Dorylus: Affordable, scalable, and accurate GNN training with distributed CPU servers and serverless threads. In: OSDI, pp. 495\u2013514. USENIX Association (2021)"},{"key":"897_CR60","doi-asserted-by":"crossref","unstructured":"Tripathy, A., Yelick, K.A., Bulu\u00e7, A.: Reducing communication in graph neural network training. In: SC, p.\u00a070. IEEE\/ACM (2020)","DOI":"10.1109\/SC41405.2020.00074"},{"key":"897_CR61","unstructured":"Velickovic, P., Cucurull, G., Casanova, A., Romero, A., Li\u00f2, P., Bengio, Y.: Graph attention networks. In: ICLR (Poster). OpenReview.net (2018)"},{"key":"897_CR62","unstructured":"Wan, B., Zhao, J., Wu, C.: Adaptive message quantization and parallelization for distributed full-graph GNN training. In: MLSys. mlsys.org (2023)"},{"key":"897_CR63","unstructured":"Wan, C., Li, Y., Li, A., Kim, N.S., Lin, Y.: BNS-GCN: efficient full-graph training of graph convolutional networks with partition-parallelism and random boundary node sampling. In: MLSys. mlsys.org (2022)"},{"key":"897_CR64","unstructured":"Wan, C., Li, Y., Wolfe, C.R., Kyrillidis, A., Kim, N.S., Lin, Y.: Pipegcn: Efficient full-graph training of graph convolutional networks with pipelined feature communication. In: ICLR. OpenReview.net (2022)"},{"key":"897_CR65","doi-asserted-by":"crossref","unstructured":"Wang, J., Yi, X., Guo, R., Jin, H., Xu, P., Li, S., Wang, X., Guo, X., Li, C., Xu, X., Yu, K., Yuan, Y., Zou, Y., Long, J., Cai, Y., Li, Z., Zhang, Z., Mo, Y., Gu, J., Jiang, R., Wei, Y., Xie, C.: Milvus: A purpose-built vector data management system. In: SIGMOD Conference, pp. 2614\u20132627. ACM (2021)","DOI":"10.1145\/3448016.3457550"},{"key":"897_CR66","unstructured":"Wang, N., Choi, J., Brand, D., Chen, C., Gopalakrishnan, K.: Training deep neural networks with 8-bit floating point numbers. In: NeurIPS, pp. 7686\u20137695 (2018)"},{"key":"897_CR67","doi-asserted-by":"publisher","unstructured":"Wang, Q., Chen, Y., Wong, W., He, B.: Hongtu: Scalable full-graph GNN training on multiple gpus. Proc. ACM Manag. Data 1(4), 246:1\u2013246:27 (2023). https:\/\/doi.org\/10.1145\/3626733","DOI":"10.1145\/3626733"},{"issue":"13","key":"897_CR68","doi-asserted-by":"publisher","first-page":"3603","DOI":"10.14778\/3424573.3424580","volume":"13","author":"R Wang","year":"2020","unstructured":"Wang, R., Deng, D.: Deltapq: Lossless product quantization code compression for high dimensional similarity search. Proc. VLDB Endow. 13(13), 3603\u20133616 (2020)","journal-title":"Proc. VLDB Endow."},{"key":"897_CR69","doi-asserted-by":"crossref","unstructured":"Wang, Z., Guan, Y., Sun, G., Niu, D., Wang, Y., Zheng, H., Han, Y.: Gnn-pim: A processing-in-memory architecture for graph neural networks. In: Conference on Advanced Computer Architecture. Springer (2020)","DOI":"10.1007\/978-981-15-8135-9_6"},{"issue":"7","key":"897_CR70","doi-asserted-by":"publisher","first-page":"807","DOI":"10.14778\/3317315.3317322","volume":"12","author":"Z Wang","year":"2019","unstructured":"Wang, Z., Kara, K., Zhang, H., Alonso, G., Zhang, C., Mutlu, O.: Accelerating generalized linear models with mlweaving: A one-size-fits-all system for any-precision learning. Proc. VLDB Endow. 12(7), 807\u2013821 (2019)","journal-title":"Proc. VLDB Endow."},{"issue":"1","key":"897_CR71","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1109\/TNNLS.2020.2978386","volume":"32","author":"Z Wu","year":"2021","unstructured":"Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., Yu, P.S.: A comprehensive survey on graph neural networks. IEEE Trans. Neural Networks Learn. Syst. 32(1), 4\u201324 (2021)","journal-title":"IEEE Trans. Neural Networks Learn. Syst."},{"key":"897_CR72","doi-asserted-by":"crossref","unstructured":"Xian, L., Li, B., Liu, J., Guo, Z., Du, D.H.C.: H-PS: A heterogeneous-aware parameter server with distributed neural network training. IEEE Access 9 (2021)","DOI":"10.1109\/ACCESS.2021.3060154"},{"key":"897_CR73","unstructured":"Yang, J.A., Park, J., Sridharan, S., Tang, P.T.P.: Training deep learning recommendation model with quantized collective communications. In: Conference on Knowledge Discovery and Data Mining (KDD) (2020)"},{"issue":"6","key":"897_CR74","doi-asserted-by":"publisher","first-page":"1241","DOI":"10.14778\/3648160.3648167","volume":"17","author":"H Yuan","year":"2024","unstructured":"Yuan, H., Liu, Y., Zhang, Y., Ai, X., Wang, Q., Chen, C., Gu, Y., Yu, G.: Comprehensive evaluation of GNN training systems: A data management perspective. Proc. VLDB Endow. 17(6), 1241\u20131254 (2024)","journal-title":"Proc. VLDB Endow."},{"key":"897_CR75","unstructured":"Zeng, H., Zhou, H., Srivastava, A., Kannan, R., Prasanna, V.K.: Graphsaint: Graph sampling based inductive learning method. In: ICLR (2020)"},{"issue":"12","key":"897_CR76","doi-asserted-by":"publisher","first-page":"3125","DOI":"10.14778\/3415478.3415539","volume":"13","author":"D Zhang","year":"2020","unstructured":"Zhang, D., Huang, X., Liu, Z., Zhou, J., Hu, Z., Song, X., Ge, Z., Wang, L., Zhang, Z., Qi, Y.: AGL: A scalable system for industrial-purpose graph machine learning. Proc. VLDB Endow. 13(12), 3125\u20133137 (2020)","journal-title":"Proc. VLDB Endow."},{"issue":"4","key":"897_CR77","doi-asserted-by":"publisher","first-page":"808","DOI":"10.14778\/3636218.3636234","volume":"17","author":"H Zhang","year":"2023","unstructured":"Zhang, H., Zhao, P., Miao, X., Shao, Y., Liu, Z., Yang, T., Cui, B.: Experimental analysis of large-scale learnable vector storage compression. Proc. VLDB Endow. 17(4), 808\u2013822 (2023)","journal-title":"Proc. VLDB Endow."},{"issue":"11","key":"897_CR78","doi-asserted-by":"publisher","first-page":"2473","DOI":"10.14778\/3476249.3476295","volume":"14","author":"W Zhang","year":"2021","unstructured":"Zhang, W., Yang, Z., Wang, Y., Shen, Y., Li, Y., Wang, L., Cui, B.: Grain: Improving data efficiency of graph neural networks via diversified influence maximization. Proc. VLDB Endow. 14(11), 2473\u20132482 (2021)","journal-title":"Proc. VLDB Endow."},{"key":"897_CR79","doi-asserted-by":"crossref","unstructured":"Zhao, K., Di, S., Dmitriev, M., Tonellot, T.D., Chen, Z., Cappello, F.: Optimizing error-bounded lossy compression for scientific data by dynamic spline interpolation. In: ICDE, pp. 1643\u20131654. IEEE (2021)","DOI":"10.1109\/ICDE51399.2021.00145"},{"key":"897_CR80","doi-asserted-by":"crossref","unstructured":"Zheng, C., Chen, H., Cheng, Y., Song, Z., Wu, Y., Li, C., Cheng, J., Yang, H., Zhang, S.: Bytegnn: efficient graph neural network training at large scale. Proceedings of the VLDB Endowment 15(6), 1228\u20131242 (2022)","DOI":"10.14778\/3514061.3514069"},{"key":"897_CR81","doi-asserted-by":"crossref","unstructured":"Zheng, D., Ma, C., Wang, M., Zhou, J., Su, Q., Song, X., Gan, Q., Zhang, Z., Karypis, G.: Distdgl: Distributed graph neural network training for billion-scale graphs. In: IA3@SC, pp. 36\u201344. IEEE (2020)","DOI":"10.1109\/IA351965.2020.00011"},{"key":"897_CR82","doi-asserted-by":"crossref","unstructured":"Zhou, H., Srivastava, A., Zeng, H., Kannan, R., Prasanna, V.K.: Accelerating large scale real-time GNN inference using channel pruning. Proc. VLDB Endow. 14(9), 1597\u20131605 (2021)","DOI":"10.14778\/3461535.3461547"},{"key":"897_CR83","doi-asserted-by":"crossref","unstructured":"Zhu, F., Gong, R., Yu, F., Liu, X., Wang, Y., Li, Z., Yang, X., Yan, J.: Towards unified INT8 training for convolutional neural network. In: CVPR, pp. 1966\u20131976. Computer Vision Foundation \/ IEEE (2020)","DOI":"10.1109\/CVPR42600.2020.00204"},{"key":"897_CR84","doi-asserted-by":"crossref","unstructured":"Zhu, R., Zhao, K., Yang, H., Lin, W., Zhou, C., Ai, B., Li, Y., Zhou, J.: Aligraph: A comprehensive graph neural network platform. Proc. VLDB Endow. 12(12) (2019)","DOI":"10.14778\/3352063.3352127"}],"container-title":["The VLDB Journal"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00778-024-00897-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00778-024-00897-2\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00778-024-00897-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,10]],"date-time":"2025-04-10T07:32:42Z","timestamp":1744270362000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00778-024-00897-2"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,1,31]]},"references-count":84,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2025,3]]}},"alternative-id":["897"],"URL":"https:\/\/doi.org\/10.1007\/s00778-024-00897-2","relation":{},"ISSN":["1066-8888","0949-877X"],"issn-type":[{"value":"1066-8888","type":"print"},{"value":"0949-877X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,1,31]]},"assertion":[{"value":"1 July 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"19 December 2024","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"28 December 2024","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"31 January 2025","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}}],"article-number":"22"}}