{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,10]],"date-time":"2026-01-10T21:12:52Z","timestamp":1768079572521,"version":"3.49.0"},"reference-count":33,"publisher":"Association for Computing Machinery (ACM)","issue":"12","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. VLDB Endow."],"published-print":{"date-parts":[[2022,8]]},"abstract":"<jats:p>Most products at ByteDance, e.g., TikTok, Douyin, and Toutiao, naturally generate massive amounts of graph data. To efficiently store, query and update massive graph data is challenging for the broad range of products at ByteDance with various performance requirements. We categorize graph workloads at ByteDance into three types: online analytical, transaction, and serving processing, where each workload has its own characteristics. Existing graph databases have different performance bottlenecks in handling these workloads and none can efficiently handle the scale of graphs at ByteDance. We developed ByteGraph to process these graph workloads with high throughput, low latency and high scalability. There are several key designs in ByteGraph that make it efficient for processing our workloads, including edge-trees to store adjacency lists for high parallelism and low memory usage, adaptive optimizations on thread pools and indexes, and geographic replications to achieve fault tolerance and availability. ByteGraph has been in production use for several years and its performance has shown to be robust for processing a wide range of graph workloads at ByteDance.<\/jats:p>","DOI":"10.14778\/3554821.3554824","type":"journal-article","created":{"date-parts":[[2022,9,29]],"date-time":"2022-09-29T22:28:39Z","timestamp":1664490519000},"page":"3306-3318","source":"Crossref","is-referenced-by-count":30,"title":["ByteGraph"],"prefix":"10.14778","volume":"15","author":[{"given":"Changji","family":"Li","sequence":"first","affiliation":[{"name":"The Chinese University of Hong Kong and ByteDance Inc"}]},{"given":"Hongzhi","family":"Chen","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Shuai","family":"Zhang","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Yingqian","family":"Hu","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Chao","family":"Chen","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Zhenjie","family":"Zhang","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Meng","family":"Li","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Xiangchen","family":"Li","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Dongqing","family":"Han","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Xiaohui","family":"Chen","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Xudong","family":"Wang","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Huiming","family":"Zhu","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Xuwei","family":"Fu","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Tingwei","family":"Wu","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Hongfei","family":"Tan","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Hengtian","family":"Ding","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Mengjin","family":"Liu","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Kangcheng","family":"Wang","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Ting","family":"Ye","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Lei","family":"Li","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Xin","family":"Li","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Yu","family":"Wang","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"Chenguang","family":"Zheng","sequence":"additional","affiliation":[{"name":"The Chinese University of Hong Kong and ByteDance Inc"}]},{"given":"Hao","family":"Yang","sequence":"additional","affiliation":[{"name":"ByteDance Inc"}]},{"given":"James","family":"Cheng","sequence":"additional","affiliation":[{"name":"The Chinese University of Hong Kong"}]}],"member":"320","published-online":{"date-parts":[[2022,9,29]]},"reference":[{"key":"e_1_2_1_1_1","unstructured":"2007. Online serving. https:\/\/cloud.google.com\/vertex-ai\/docs\/featurestore\/serving-online.  2007. Online serving. https:\/\/cloud.google.com\/vertex-ai\/docs\/featurestore\/serving-online."},{"key":"e_1_2_1_2_1","unstructured":"2015. Titan. https:\/\/titan.thinkaurelius.com\/.  2015. Titan. https:\/\/titan.thinkaurelius.com\/."},{"key":"e_1_2_1_3_1","unstructured":"2021. AgensGraph. https:\/\/bitnine.net\/.  2021. AgensGraph. https:\/\/bitnine.net\/."},{"key":"e_1_2_1_4_1","unstructured":"2021. ArangoDB. https:\/\/www.arangodb.com\/.  2021. ArangoDB. https:\/\/www.arangodb.com\/."},{"key":"e_1_2_1_5_1","unstructured":"2021. JanusGraph. https:\/\/janusgraph.org\/.  2021. JanusGraph. https:\/\/janusgraph.org\/."},{"key":"e_1_2_1_6_1","unstructured":"2022. Alibaba GDB. https:\/\/www.aliyun.com\/product\/gdb\/.  2022. Alibaba GDB. https:\/\/www.aliyun.com\/product\/gdb\/."},{"key":"e_1_2_1_7_1","unstructured":"2022. Apache TinkerPop: Gremlin Query Language. https:\/\/tinkerpop.apache.org\/gremlin.html.  2022. Apache TinkerPop: Gremlin Query Language. https:\/\/tinkerpop.apache.org\/gremlin.html."},{"key":"e_1_2_1_8_1","unstructured":"2022. AWS Neptune. https:\/\/aws.amazon.com\/neptune\/.  2022. AWS Neptune. https:\/\/aws.amazon.com\/neptune\/."},{"key":"e_1_2_1_9_1","unstructured":"2022. Azure Cosmos DB. https:\/\/docs.microsoft.com\/en-us\/azure\/cosmos-db\/graph\/graph-introduction.  2022. Azure Cosmos DB. https:\/\/docs.microsoft.com\/en-us\/azure\/cosmos-db\/graph\/graph-introduction."},{"key":"e_1_2_1_10_1","unstructured":"2022. Neo4j. https:\/\/neo4j.com\/.  2022. Neo4j. https:\/\/neo4j.com\/."},{"key":"e_1_2_1_11_1","unstructured":"2022. PostgreSQL. https:\/\/www.postgresql.org\/.  2022. PostgreSQL. https:\/\/www.postgresql.org\/."},{"key":"e_1_2_1_12_1","unstructured":"2022. RocksDB. http:\/\/rocksdb.org\/.  2022. RocksDB. http:\/\/rocksdb.org\/."},{"key":"e_1_2_1_13_1","volume-title":"MySQL High Availability - Tools for Building Robust Data Centers","author":"Bell Charles","year":"2002","unstructured":"Charles Bell , Mats Kindahl , and Lars Thalmann . 2014. MySQL High Availability - Tools for Building Robust Data Centers , 2 nd Edition. O'Reilly . http:\/\/shop.oreilly.com\/product\/06369 2002 6907.do Charles Bell, Mats Kindahl, and Lars Thalmann. 2014. MySQL High Availability - Tools for Building Robust Data Centers, 2nd Edition. O'Reilly. http:\/\/shop.oreilly.com\/product\/0636920026907.do","edition":"2"},{"key":"e_1_2_1_14_1","volume-title":"2013 USENIX Annual Technical Conference","author":"Bronson Nathan","year":"2013","unstructured":"Nathan Bronson , Zach Amsden , George Cabrera , Prasad Chakka , Peter Dimov , Hui Ding , Jack Ferris , Anthony Giardullo , Sachin Kulkarni , Harry C. Li , Mark Marchukov , Dmitri Petrov , Lovro Puzar , Yee Jiun Song , and Venkateshwaran Venkataramani . 2013 . TAO: Facebook's Distributed Data Store for the Social Graph . In 2013 USENIX Annual Technical Conference , San Jose, CA, USA, June 26--28 , 2013. USENIX Association, 49--60. https:\/\/www.usenix.org\/conference\/atc13\/technical-sessions\/presentation\/bronson Nathan Bronson, Zach Amsden, George Cabrera, Prasad Chakka, Peter Dimov, Hui Ding, Jack Ferris, Anthony Giardullo, Sachin Kulkarni, Harry C. Li, Mark Marchukov, Dmitri Petrov, Lovro Puzar, Yee Jiun Song, and Venkateshwaran Venkataramani. 2013. TAO: Facebook's Distributed Data Store for the Social Graph. In 2013 USENIX Annual Technical Conference, San Jose, CA, USA, June 26--28, 2013. USENIX Association, 49--60. https:\/\/www.usenix.org\/conference\/atc13\/technical-sessions\/presentation\/bronson"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3386135"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1145\/3357223.3362715"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.14778\/3551793.3551813"},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3384685"},{"key":"e_1_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.14778\/3476311.3476379"},{"key":"e_1_2_1_20_1","volume-title":"Lee","author":"Deutsch Alin","year":"2019","unstructured":"Alin Deutsch , Yu Xu , Mingxi Wu , and Victor E . Lee . 2019 . TigerGraph: A Native MPP Graph Database. CoRR abs\/1901.08248 (2019). arXiv:1901.08248 http:\/\/arxiv.org\/abs\/1901.08248 Alin Deutsch, Yu Xu, Mingxi Wu, and Victor E. Lee. 2019. TigerGraph: A Native MPP Graph Database. CoRR abs\/1901.08248 (2019). arXiv:1901.08248 http:\/\/arxiv.org\/abs\/1901.08248"},{"key":"e_1_2_1_21_1","volume-title":"Proceedings of the 11th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2014","author":"Dragojevic Aleksandar","year":"2014","unstructured":"Aleksandar Dragojevic , Dushyanth Narayanan , Miguel Castro , and Orion Hodson . 2014 . FaRM: Fast Remote Memory . In Proceedings of the 11th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2014 , Seattle, WA, USA, April 2--4 , 2014. USENIX Association, 401--414. https:\/\/www.usenix.org\/conference\/nsdi14\/technical-sessions\/dragojevi%C4%87 Aleksandar Dragojevic, Dushyanth Narayanan, Miguel Castro, and Orion Hodson. 2014. FaRM: Fast Remote Memory. In Proceedings of the 11th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2014, Seattle, WA, USA, April 2--4, 2014. USENIX Association, 401--414. https:\/\/www.usenix.org\/conference\/nsdi14\/technical-sessions\/dragojevi%C4%87"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1145\/2815400.2815425"},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.14778\/2983200.2983202"},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.14778\/3476249.3476297"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.14778\/3415478.3415550"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1145\/3035918.3056445"},{"key":"e_1_2_1_27_1","doi-asserted-by":"publisher","DOI":"10.1145\/258533.258660"},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-14472-6_2"},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1145\/3459637.3481913"},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.14778\/3342263.3342643"},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1137\/1.9781611975031.39"},{"key":"e_1_2_1_32_1","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3386134"},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.14778\/3384345.3384351"}],"container-title":["Proceedings of the VLDB Endowment"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.14778\/3554821.3554824","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,28]],"date-time":"2022-12-28T11:21:35Z","timestamp":1672226495000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.14778\/3554821.3554824"}},"subtitle":["a high-performance distributed graph database in ByteDance"],"short-title":[],"issued":{"date-parts":[[2022,8]]},"references-count":33,"journal-issue":{"issue":"12","published-print":{"date-parts":[[2022,8]]}},"alternative-id":["10.14778\/3554821.3554824"],"URL":"https:\/\/doi.org\/10.14778\/3554821.3554824","relation":{},"ISSN":["2150-8097"],"issn-type":[{"value":"2150-8097","type":"print"}],"subject":[],"published":{"date-parts":[[2022,8]]}}}