max-background-jobs: 8 raftdb.max-background-jobs: 4 raftdb.allow-concurrent-memtable-write: true server.grpc-concurrency: 6 readpool.unified.min-thread-count: 5 readpool.unified.max-thread-count: 20 readpool unified.max-thread-count: 20 readpool.unified.min-thread-count: 5 rocksdb.max-background-jobs: 8 server.grpc-concurrency: 6 storage.scheduler-worker-pool-size: 20 server.enable-request-batch: false TiDB global parameters • The gRPC compression format (none by default): To increase the transmission speed of gRPC packages between geo-distributed target nodes, set this parameter to gzip. server.grpc-compression-type:

expressions with side effects in column pruning and aggregation pushdown #27106 – Remove useless gRPC logs #24190 – Limit the valid decimal length to fix precision-related issues #3091 – Fix the issue max-background-jobs: 8 raftdb.max-background-jobs: 4 raftdb.allow-concurrent-memtable-write: true server.grpc-concurrency: 6 readpool.unified.min-thread-count: 5 readpool.unified.max-thread-count: 20 readpool unified.max-thread-count: 20 readpool.unified.min-thread-count: 5 rocksdb.max-background-jobs: 8 server.grpc-concurrency: 6 storage.scheduler-worker-pool-size: 20 TiDB global variable configuration set global

max-background-jobs: 8 raftdb.max-background-jobs: 4 raftdb.allow-concurrent-memtable-write: true server.grpc-concurrency: 6 readpool.unified.min-thread-count: 5 50 readpool.unified.max-thread-count: 20 readpool unified.max-thread-count: 20 readpool.unified.min-thread-count: 5 rocksdb.max-background-jobs: 8 server.grpc-concurrency: 6 storage.scheduler-worker-pool-size: 20 server.enable-request-batch: false TiDB global parameters • The gRPC compression format (none by default): To increase the transmission speed of gRPC packages between geo-distributed target nodes, set this parameter to gzip. server.grpc-compression-type:

TiFlash – Optimize the communication of local operators – Increase the non-temporary thread count of gRPC to avoid the frequent creation or destruction of threads 53 • Tools – Backup & Restore (BR) * max-background-jobs: 8 raftdb.max-background-jobs: 4 raftdb.allow-concurrent-memtable-write: true server.grpc-concurrency: 6 readpool.unified.min-thread-count: 5 readpool.unified.max-thread-count: 20 readpool unified.max-thread-count: 20 readpool.unified.min-thread-count: 5 rocksdb.max-background-jobs: 8 server.grpc-concurrency: 6 storage.scheduler-worker-pool-size: 20 TiDB global variable configuration set global

Description TiDB grpc- �→ keepalive �→ - �→ time Modified Adds the mini- mum value of 1. 68 Configuration file or compo- nent Configuration parame- ter Change type Description TiDB grpc- �→ keepalive parameters • The gRPC compression format (none by default): To increase the transmission speed of gRPC packages between geo-distributed target nodes, set this parameter to gzip. server.grpc-compression-type: The maximum concurrency limits for other modules on TiKV, such as storage readpool, coprocessor, and gRPC, can be adjusted through the TiKV configuration file. The actual CPU usage can be observed through

max-background-jobs: 8 raftdb.max-background-jobs: 4 raftdb.allow-concurrent-memtable-write: true server.grpc-concurrency: 6 readpool.unified.min-thread-count: 5 readpool.unified.max-thread-count: 20 readpool max-thread-count: 20 readpool.unified.min-thread-count: 5 rocksdb.max-background-jobs: 8 server.grpc-concurrency: 6 storage.scheduler-worker-pool-size: 20 server.enable-request-batch: false TiDB 全局变量配置 4.3.6.1.2 关键参数配置 本节介绍跨数据中心部署 TiDB 集群的关键参数配置。 TiKV 参数 • 设置 gRPC 的压缩格式，默认为 none。为提高跨机房部署场景的目标节点间 gRPC 包的传输速度，建议设 置为 gzip 格式。 server.grpc-compression-type: gzip • label 配置 由于采用跨机房部署 TiKV，为了避免物理机宕机导致

max-background-jobs: 8 raftdb.max-background-jobs: 4 raftdb.allow-concurrent-memtable-write: true server.grpc-concurrency: 6 readpool.unified.min-thread-count: 5 readpool.unified.max-thread-count: 20 readpool max-thread-count: 20 readpool.unified.min-thread-count: 5 rocksdb.max-background-jobs: 8 server.grpc-concurrency: 6 storage.scheduler-worker-pool-size: 20 server.enable-request-batch: false TiDB 全局变量配置 4.3.6.1.2 关键参数配置 本节介绍跨数据中心部署 TiDB 集群的关键参数配置。 TiKV 参数 • 设置 gRPC 的压缩格式，默认为 none。为提高跨机房部署场景的目标节点间 gRPC 包的传输速度，建议设 置为 gzip 格式。 server.grpc-compression-type: gzip • label 配置 由于采用跨机房部署 TiKV，为了避免物理机宕机导致

foreground-write-bandwidth • quota.foreground-read-bandwidth • quota.max-delay-duration • server.grpc-memory-pool-quota • server.max-grpc-send-msg-len • server.raft-msg-max-batch-size • In v6.1.0, some configuration file blocking Raftstore message loops #11966 • Support dynamic setting of the the maximum message length (max-grpc-send �→ -msg-len) and the maximum batch size of gPRC messages (raft-msg-max- �→ batch-size) #12334 5 raftstore.store-pool-size: 3 raftstore.apply-pool-size: 3 rocksdb.max-background-jobs: 8 server.grpc-concurrency: 6 readpool.storage.normal-concurrency: 10 TiDB global variable configuration set global

default. 51 Configuration file Configuration parame- ter Change type Description TiKV server. �→ grpc �→ - �→ compression �→ - �→ type Modified This configu- ration item now also controls the com- pression that the gRPC request sources duration metric is displayed incorrectly in the monitoring dashboard #17133 @King-Dylan • Fix the issue that setting the gRPC message compression method via grpc- �→ compression-type parameters • The gRPC compression format (none by default): To increase the transmission speed of gRPC packages between geo-distributed target nodes, set this parameter to gzip. server.grpc-compression-type:

Description TiDB grpc- �→ keepalive �→ - �→ time Modified Adds the mini- mum value of 1. 63 Configuration file or compo- nent Configuration parame- ter Change type Description TiDB grpc- �→ keepalive parameters • The gRPC compression format (none by default): To increase the transmission speed of gRPC packages between geo-distributed target nodes, set this parameter to gzip. server.grpc-compression-type: The maximum concurrency limits for other modules on TiKV, such as storage readpool, coprocessor, and gRPC, can be adjusted through the TiKV configuration file. The actual CPU usage can be observed through