erigon-pulse/cmd/rpctest

Create files with sample requests

go run ./cmd/rpctest/main.go bench1 will print tmpDir. And create in tmpDir results_*.csv and vegeta_*.txt files.

Command takes long time. Kill it when vegeta_turbo_geth_debug_storageRangeAt.txt is few MB.

File vegeta_turbo_geth_*.txt will produce load to turbo_geth node, vegeta_geth_*.txt to geth. Change host/port in --gethUrl, --tgUrl variable.

By default go run ./cmd/rpctest/main.go bench1 calling only turbogeth node because cmd/rpctest/rpctest/bench1.go calling it with first parameter needCompare=false. Set --needCompare to call Geth and TurboGeth nodes and compare results.

Install Vegeta

go get -u github.com/tsenart/vegeta

Run vegeta

tmpDir = "/var/folders/x_/1mnbt25s3291zr5_fxhjfnq9n86kng/T/"
cat $(tmpDir)/turbo_geth_stress_test/vegeta_geth_debug_storageRangeAt.csv | vegeta attack -rate=600 -format=json -duration=20s -timeout=300s | vegeta plot > plot.html
open plot.html

Mac environment changes

Change Open Files Limit

Results from my Macbook:

start rpcdaemon with turbo_geth:

GODEBUG=remotedb.debug=1 go run ./cmd/tg --private.api.addr localhost:9997   --rpcport 8545  --rpc --rpcapi eth,debug,net --nodiscover
GODEBUG=remotedb.debug=1 go run ./cmd/rpcdaemon --rpcapi eth,debug,net --rpcport 9545 --private.api.addr 127.0.0.1:9997

On simple requests eth_getBlockByNumber RPC Daemon looks well:

cat /tmp/turbo_geth_stress_test/vegeta_turbo_geth_eth_getBlockByNumber.txt | vegeta attack -rate=1000 -format=json -duration=20s -timeout=300s | vegeta report

300rps: 
- Geth Alone: 80% of CPU, 95-Latency 2ms

- Geth Behind RPC Daemon: 25% of CPU
- RPC Daemon: 45% of CPU, 95-Latency 3ms

1000rps: 
- Geth Alone: 200% of CPU, 95-Latency 3ms

- Geth Behind RPC Daemon: 50% of CPU
- RPC Daemon: 120% of CPU, 95-Latency 6ms

2000rps: 
- Geth Alone: 400% of CPU, 95-Latency 15ms

- Geth Behind RPC Daemon: 100% of CPU
- RPC Daemon: 250% of CPU, 95-Latency 10ms

On complex request - debug_storageRangeAt producing >600 db.View calls with twice more .Bucket/.Cursor/.Seek calls:

echo "POST http://localhost:8545 \n Content-Type: application/json \n @$(pwd)/cmd/rpctest/heavyStorageRangeAt.json" | vegeta attack -rate=20 -duration=20s -timeout=300s | vegeta report

10rps, batchSize 10K:
- Geth Alone: 100% of CPU, 95-Latency 15ms 

- Geth Behind RPC Daemon: 200% of CPU
- RPC Daemon: 230% of CPU, 95-Latency 7s

10rps, batchSize 10:
- Geth Alone: 100% of CPU, 95-Latency 15ms 

- Geth Behind RPC Daemon: 110% of CPU
- RPC Daemon: 100% of CPU, 95-Latency 230ms

Reason is: often usage of .GetAsOf() - this method does much .Next() and .Seek() calls. Each .Seek() call invalidate internal batch cache of .Next() method and remote_db does read CursorBatchSize amount of keys again.

PoolSize=128, CursorBatchSize=10K -> 95-Latency 30s (eat all conns in pool)
PoolSize=128, CursorBatchSize=1K -> 95-Latency 6s (eat 50 conns in pool)
PoolSize=128, CursorBatchSize=100 -> 95-Latency 600ms (eat 5 conns in pool)

Idea to discuss: implement CmdGetAsOf

BenchmarkSerialize:

BenchmarkSerialize/encodeKeyValue()-12         	 4249026	       268 ns/op	      12 B/op	       0 allocs/op
BenchmarkSerialize/encoder.Encode(&k)-12       	 4702418	       258 ns/op	      14 B/op	       0 allocs/op
BenchmarkSerialize/encoder.Encode(k)-12        	 3382797	       350 ns/op	     104 B/op	       2 allocs/op
BenchmarkSerialize/encoder.MustEncode(&k)-12   	 8431810	       140 ns/op	       0 B/op	       0 allocs/op
BenchmarkSerialize/encoder.MustEncode(k)-12    	 5446293	       262 ns/op	     114 B/op	       2 allocs/op
BenchmarkSerialize/Encode(struct)-12           	 4160940	       266 ns/op	       0 B/op	       0 allocs/op
BenchmarkSerialize/10K_Encode(&k,_&v)-12       	    1368	   1089648 ns/op	  402976 B/op	       0 allocs/op
BenchmarkSerialize/Encode([10K]k,_[10K]v)-12   	    1825	    548953 ns/op	  491584 B/op	       4 allocs/op