Sharded state experiment based on call_traces stage run via integration (#1329)
* Initial work on state shards * Fix compile errors * Intermediate * Implement dispatcher * Implemented sharded reader * No cache * Print comms * No prepopulation * Print dispatcher activity * Print more * Print errors * Fix dispatcher * Remove logging * Dispatcher to enforce initial barrier * Don't print that much * Turn off cache pre-seeting for state sharding * Artificial latency, fixed number of blocks to execute * Fix
Showing
- Makefile 1 addition, 0 deletionsMakefile
- cmd/headers/download/sentry.go 1 addition, 1 deletioncmd/headers/download/sentry.go
- cmd/integration/commands/flags.go 16 additions, 0 deletionscmd/integration/commands/flags.go
- cmd/integration/commands/stages.go 185 additions, 1 deletioncmd/integration/commands/stages.go
- eth/stagedsync/stage_call_traces.go 52 additions, 33 deletionseth/stagedsync/stage_call_traces.go
- eth/stagedsync/stagebuilder.go 2 additions, 1 deletioneth/stagedsync/stagebuilder.go
- turbo/shards/shards.go 355 additions, 0 deletionsturbo/shards/shards.go
- turbo/shards/shards.pb.go 162 additions, 0 deletionsturbo/shards/shards.pb.go
- turbo/shards/shards.proto 15 additions, 0 deletionsturbo/shards/shards.proto
- turbo/shards/shards_grpc.pb.go 129 additions, 0 deletionsturbo/shards/shards_grpc.pb.go
- turbo/shards/shards_test.go 9 additions, 0 deletionsturbo/shards/shards_test.go
Loading
Please register or sign in to comment