ByteDance distributed graph database based on brpc in practice


Chinese Session 2021-08-06 14:10 GMT+8  #incubator

ByteGraph is a distributed graph database developed by ByteDance, which supports storing trillions of edges of data scale of very large graphs, supports ten million read and write QPS, and runs on a scale of nearly 10,000 servers. In such a large-scale scenario, ByteGraph relies on brpc to provide high-performance, easy-to-use and stable rpc framework support. This talk will introduce ByteGraph and brpc respectively, and the practical experience of a large-scale distributed system like ByteGraph based on brpc.


Lei Li: Graduated from Xi’an Jiaotong University, he is currently working on infrastructure research and development at ByteDance, focusing on graph databases, graph computing, graph neural networks, and distributed KV systems.