{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,20]],"date-time":"2025-12-20T08:39:50Z","timestamp":1766219990546,"version":"3.48.0"},"publisher-location":"New York, NY, USA","reference-count":39,"publisher":"ACM","content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2025,9,8]]},"DOI":"10.1145\/3754598.3754643","type":"proceedings-article","created":{"date-parts":[[2025,12,20]],"date-time":"2025-12-20T08:34:32Z","timestamp":1766219672000},"page":"53-62","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["A High-Accuracy Sketch for Measuring Low-Entropy Flows in Distributed AI Training"],"prefix":"10.1145","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5473-8738","authenticated-orcid":false,"given":"Jin","family":"Wang","sequence":"first","affiliation":[{"name":"Sanya Institute of Hunan University of Science and Technology, Sanya,Hainan, China"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-9674-0523","authenticated-orcid":false,"given":"Chenye","family":"Zhu","sequence":"additional","affiliation":[{"name":"Hunan University of Science and Technology, Xiangtan,Hunan, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8216-9683","authenticated-orcid":false,"given":"Jinbin","family":"Hu","sequence":"additional","affiliation":[{"name":"Changsha University of Science and Technology, Changsha,Hunan, China"}]}],"member":"320","published-online":{"date-parts":[[2025,12,20]]},"reference":[{"key":"e_1_3_3_2_2_2","doi-asserted-by":"crossref","unstructured":"H. Lim and J. Ye and J. S. Abdu and D. Han and ohthers. Accelerating model training in multi-cluster environments with consumer-grade gpus. In Proc. ACM SIGCOMM 2024.","DOI":"10.1145\/3651890.3672228"},{"key":"e_1_3_3_2_3_2","doi-asserted-by":"crossref","unstructured":"A. Gangidi and R. Miao and S. Zheng and S. J. Bondu and G. Goes and H. Morsy and R. Puri and M. Riftadi and A. J. Shetty and J. Yang. Rdma over ethernet for distributed training at meta scale. In Proc. ACM SIGCOMM 2024.","DOI":"10.1145\/3651890.3672233"},{"key":"e_1_3_3_2_4_2","doi-asserted-by":"crossref","unstructured":"H. Luo and J. Zhang and M. Yu and Y. Pan and T. Pan and T. Huang. Network load balancing with no out-of-order packet for RoCE. IEEE Network 2024.","DOI":"10.1109\/MNET.2024.3503714"},{"key":"e_1_3_3_2_5_2","doi-asserted-by":"crossref","unstructured":"J. Han and K. Xue and W. Wang and R. Li and Q. Sun and J. Lu. RateMP: Optimizing Bandwidth Utilization with High Burst Tolerance in Data Center Networks. In Proc. IEEE INFOCOM 2024.","DOI":"10.1109\/INFOCOM52122.2024.10621096"},{"key":"e_1_3_3_2_6_2","doi-asserted-by":"crossref","unstructured":"S. Li and J. Huang and W. Zhang and J. Shao. PA-Sketch: a fast and accurate sketch for differentiated flow estimation. In Proc. IEEE ICNP 2023.","DOI":"10.1109\/ICNP59255.2023.10355581"},{"key":"e_1_3_3_2_7_2","doi-asserted-by":"crossref","unstructured":"C. Estan and G. Varghese. New directions in traffic measurement and accounting. In Proc. ACM SIGCOMM 2002.","DOI":"10.1145\/505208.505212"},{"key":"e_1_3_3_2_8_2","unstructured":"X. Li and Z. Fan and H. Li and Z. Zhong and J. Guo and S. Long and T. Yang and B. Cui. In Proc. IEEE\/ACM IWQoS 2023."},{"key":"e_1_3_3_2_9_2","doi-asserted-by":"crossref","unstructured":"J. He and J. Zhu and Q. Huang. Histsketch: A compact data structure for accurate per-key distribution monitoring. In Proc. IEEE ICDE 2023.","DOI":"10.1109\/ICDE55515.2023.00156"},{"key":"e_1_3_3_2_10_2","doi-asserted-by":"crossref","unstructured":"K. C. Lin and W. Lai. Mc-sketch: Enabling heterogeneous network monitoring resolutions with multi-class sketch. In Proc. IEEE INFOCOM 2022.","DOI":"10.1109\/INFOCOM48880.2022.9796955"},{"key":"e_1_3_3_2_11_2","doi-asserted-by":"crossref","unstructured":"Z. Liu and Y. Zhao and Z. Fan and T. Yang and X. Li and R. Zhang and K. Yang and Z. Jiang and Z. Zhong and Y. Huang. Burstbalancer: Do less better balance for large-scale data center traffic. IEEE Transactions on Parallel and Distributed Systems 35(6):932\u2013949 2023.","DOI":"10.1109\/TPDS.2023.3295454"},{"key":"e_1_3_3_2_12_2","unstructured":"S. Rajasekaran and M. Ghobadi and A. Akella. { CASSINI} :{ Network-Aware} Job Scheduling in Machine Learning Clusters. In Proc. USENIX NSDI 2024."},{"key":"e_1_3_3_2_13_2","unstructured":"H. Wang and H. Tian and J. Chen and X. Wan and J. Xia and G. Zeng and W. Bai and J. Jiang and Y. Wang and K. Chen. Towards { Domain-Specific} network transport for distributed { DNN} training. In Proc. USENIX NSDI 2024."},{"key":"e_1_3_3_2_14_2","doi-asserted-by":"crossref","unstructured":"J. Jiang and F. Fu and T. Yang and B. Cui. Sketchml: Accelerating distributed machine learning with data sketches. In Proc. ACM SIGMOD 2018.","DOI":"10.1145\/3183713.3196894"},{"key":"e_1_3_3_2_15_2","doi-asserted-by":"crossref","unstructured":"T. Yang and J. Li and Y. Zhao and K. Yang and J. Jiang and Y. Zhang and N. Zhang. Qcluster: Clustering packets for flow scheduling. In Proc. ACM WWW 2022.","DOI":"10.1145\/3485447.3511980"},{"key":"e_1_3_3_2_16_2","doi-asserted-by":"crossref","unstructured":"Q. Wen and K. He and L. Sun and Y. Zhang and M. Ke and H. Xu. RobustPeriod: Robust time-frequency mining for multiple periodicity detection. In Proc. ACM SIGMOD 2021.","DOI":"10.1145\/3448016.3452779"},{"key":"e_1_3_3_2_17_2","doi-asserted-by":"crossref","unstructured":"Z. Liu and C. Kong and K. Yang and T. Yang and R. Miao and Q. Chen and Y. Zhao and Y. Tu and B. Cui. Hypercalm sketch: One-pass mining periodic batches in data streams. In Proc. IEEE ICDE 2023.","DOI":"10.1109\/ICDE55515.2023.00009"},{"key":"e_1_3_3_2_18_2","doi-asserted-by":"crossref","unstructured":"Z. Fan and Y. Zhang and T. Yang and M. Yan and G. Wen and Y. Wu and B. Cui. Periodicsketch: Finding periodic items in data streams. In Proc. IEEE ICDE 2022.","DOI":"10.1109\/ICDE53745.2022.00012"},{"key":"e_1_3_3_2_19_2","doi-asserted-by":"crossref","unstructured":"P. Chen and D. Chen and L. Zheng and J. Li and T. Yang. Out of many we are one: Measuring item batch with clock-sketch. In Proc. ACM SIGMOD 2021.","DOI":"10.1145\/3448016.3452784"},{"key":"e_1_3_3_2_20_2","doi-asserted-by":"crossref","unstructured":"K. Yang and S. Long and Q. Shi and Y. Li and T. Yang and Z. Jia. SketchINT: Empowering INT with TowerSketch for per-flow per-switch measurement. IEEE Transactions on Parallel and Distributed Systems 34(11): 2876\u20132894 2023.","DOI":"10.1109\/TPDS.2023.3303924"},{"key":"e_1_3_3_2_21_2","doi-asserted-by":"crossref","unstructured":"R. Miao and F. Dong and Y. Zhao and Y. Zhao and T. Yang and B. Cui. Sketchconf: A framework for automatic sketch configuration. In Proc. IEEE ICDE 2023.","DOI":"10.1109\/ICDE55515.2023.00157"},{"key":"e_1_3_3_2_22_2","doi-asserted-by":"crossref","unstructured":"Y. Zhao and Y. Zhang and P. Yi and T. Yang and B. Cui. The stair sketch: Bringing more clarity to memorize recent events. In Proc. IEEE ICDE 2022.","DOI":"10.1109\/ICDE53745.2022.00017"},{"key":"e_1_3_3_2_23_2","doi-asserted-by":"crossref","unstructured":"R. B. Basat and G. Einziger and M. Mitzenmacher and S. Vargaftik. Salsa: self-adjusting lean streaming analytics. IEEE ICDE 2021.","DOI":"10.1109\/ICDE51399.2021.00080"},{"key":"e_1_3_3_2_24_2","doi-asserted-by":"crossref","unstructured":"Q. Chen and Y. Hong and Y. Wu and T. Yang and B. Cui. CodingSketch: A Hierarchical Sketch with Efficient Encoding and Recursive Decoding. In Proc. IEEE ICDE 2024.","DOI":"10.1109\/ICDE60146.2024.00130"},{"key":"e_1_3_3_2_25_2","doi-asserted-by":"crossref","unstructured":"T. Yang and J. Jiang and P. Liu and Q. Huang and J.Gong and Y. Zhou and R.Miao and X. Li and S.Uhlig. Elastic sketch: Adaptive and fast network-wide measurements. In Proc. ACM SIGCOMM 2018.","DOI":"10.1145\/3230543.3230544"},{"key":"e_1_3_3_2_26_2","doi-asserted-by":"crossref","unstructured":"T. Yang and J. Gong and H. Zhang and L. Zou and L. Shi and X. Li. Heavyguardian: Separate and guard hot items in data streams. In Proc. ACM SIGKDD 2018.","DOI":"10.1145\/3219819.3219978"},{"key":"e_1_3_3_2_27_2","doi-asserted-by":"crossref","unstructured":"T. Yang and H. Zhang and J. Li and J. Gong and S. Uhlig and S. Chen and X. Li. HeavyKeeper: an accurate algorithm for finding Top-k elephant flows. IEEE\/ACM Transactions on Networking 27(5): 1845\u20131858 2019.","DOI":"10.1109\/TNET.2019.2933868"},{"key":"e_1_3_3_2_28_2","doi-asserted-by":"crossref","unstructured":"Q. Shi and Y. Xu and J. Qi and W. Li and T. Yang and Y. Xu and Y. Wang. Cuckoo counter: Adaptive structure of counters for accurate frequency and top-k estimation. IEEE\/ACM Transactions on Networking 31(4):1854\u20131869 2023.","DOI":"10.1109\/TNET.2022.3232098"},{"key":"e_1_3_3_2_29_2","doi-asserted-by":"crossref","unstructured":"T. Yang and Y. Zhou and H. Jin and X. Li. Pyramid sketch: A sketch framework for frequency estimation of data streams. In Proc. VLDB Endowment 2016.","DOI":"10.14778\/3137628.3137652"},{"key":"e_1_3_3_2_30_2","doi-asserted-by":"crossref","unstructured":"T. Yang and S. Gao and Z. Sun and Y. Wang and Y. Shen and X. Li. Diamond sketch: Accurate per-flow measurement for big streaming data. IEEE Transactions on Parallel and Distributed Systems 30(12): 2650\u20132662 2019.","DOI":"10.1109\/TPDS.2019.2923772"},{"key":"e_1_3_3_2_31_2","doi-asserted-by":"crossref","unstructured":"H. Li and Q. Cheni and T. Yang and B. Cui. Stingy sketch: A sketch framework for accurate and fast frequency estimation. In Proc. VLDB Endowment 2022.","DOI":"10.14778\/3523210.3523220"},{"key":"e_1_3_3_2_32_2","doi-asserted-by":"crossref","unstructured":"G. Cormode and S. Muthukrishnan. An improved data stream summary: the count-min sketch and its applications. Journal of Algorithms 55(1): 58\u201375 2005.","DOI":"10.1016\/j.jalgor.2003.12.001"},{"key":"e_1_3_3_2_33_2","doi-asserted-by":"crossref","unstructured":"Y. Li and F. Wang and X. Yu and Y. Yang and K. Yang and T. Yang and Z. Ma and B. Cui and S. Uhlig. Ladderfilter: Filtering infrequent items with small memory and time overhead. In Proc. ACM SIGMOD 2023.","DOI":"10.1145\/3588690"},{"key":"e_1_3_3_2_34_2","doi-asserted-by":"crossref","unstructured":"Y. Zhao and W. Han and Z. Zhong and Y. Zhang and T. Yang and B. Cui. Double-anonymous sketch: Achieving top-k-fairness for finding global top-k frequent items. In Proc. ACM SIGMOD 2023.","DOI":"10.1145\/3588933"},{"key":"e_1_3_3_2_35_2","doi-asserted-by":"crossref","unstructured":"J. Li and Z. Li and Y. Xu and S. Jiang and T. Yang and B. Cui and Y. Dai and G. Zhang. Wavingsketch: An unbiased and generic sketch for finding top-k items in data streams. In Proc. ACM SIGKDD 2020.","DOI":"10.1145\/3394486.3403208"},{"key":"e_1_3_3_2_36_2","doi-asserted-by":"crossref","unstructured":"X. Gou and L. He and Y. Zhang and K. Wang and X. Liu and T. Yang and Y. Wang and B. Cui. Sliding sketches: A framework using time zones for data stream processing in sliding windows. In Proc. ACM SIGKDD 2020.","DOI":"10.1145\/3394486.3403144"},{"key":"e_1_3_3_2_37_2","doi-asserted-by":"crossref","unstructured":"J. Huang and W. Zhang and Y. Li and L. Li and Z. Li and J. Ye and J. Wang. ChainSketch: An efficient and accurate sketch for heavy flow detection. IEEE\/ACM Transactions on Networking 31(2): 738\u2013753 2022.","DOI":"10.1109\/TNET.2022.3199506"},{"key":"e_1_3_3_2_38_2","doi-asserted-by":"crossref","unstructured":"R. Ben-Basat and G. Einziger and R. Friedman and Y. Kassner. Heavy hitters in streams and sliding windows. In Proc. IEEE INFOCOM 2016.","DOI":"10.1109\/INFOCOM.2016.7524364"},{"key":"e_1_3_3_2_39_2","unstructured":"https:\/\/www.caida.org\/funding\/."},{"key":"e_1_3_3_2_40_2","doi-asserted-by":"crossref","unstructured":"Z. Liu and Y. Zhang and Y. Zhu and R. Zhang and T. Yang and K. Xie and S. Wang and T. Li and B. Cui. Treesensing: Linearly compressing sketches with flexibility. In Proc. ACM SIGMOD 2023.","DOI":"10.1145\/3588910"}],"event":{"name":"ICPP '25: 54th International Conference on Parallel Processing","location":"San Diego CA USA","acronym":"ICPP '25"},"container-title":["Proceedings of the 54th International Conference on Parallel Processing"],"original-title":[],"link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3754598.3754643","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,20]],"date-time":"2025-12-20T08:37:11Z","timestamp":1766219831000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3754598.3754643"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,8]]},"references-count":39,"alternative-id":["10.1145\/3754598.3754643","10.1145\/3754598"],"URL":"https:\/\/doi.org\/10.1145\/3754598.3754643","relation":{},"subject":[],"published":{"date-parts":[[2025,9,8]]},"assertion":[{"value":"2025-12-20","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}