About performance between ByConity0.3.2 and Clickhouse 23.1.3

[merseyalma]

I made a performance comparison between ByConity0.3.2 and Clickhouse 23.1.3.

It contains two items.

• a csv file has 104 fields and 2.15M records, import it to a table named A in the two databases;
• execute a sql which stats A in the two databases;

the result:

database	hardware	import	execution
ClickHouse server 23.1.3	16G Memery 8 Core 2.1Ghz CPU	16s	3.9s
ByConity0.3.2	16G Memery 8 Core 2.1Ghz CPU	20s	6.5s

Could you help me to explain why the import and execution in ByConity is slower than Clickhouse, and help me to optimize ? If you can , contact me by Webchat ( id:merseyalma) to get more details about data and table denifition.

[kevinthfang]

1. Could you please explain how you deploy byconity? E.g. Are you using HDFS or S3? If hdfs, how the HDFS is deployed?
2. Is the execution 1st time query?

[merseyalma]

1 deploy by docker-compose. I only changed .env file as below

COMPOSE_PROJECT_NAME="random"

DATA_DIR=./data
SERVER_TCP_PORT=9123
SERVER_HTTP_PORT=8123

#BYCONITY_IMAGE=byconity/debian-runner
BYCONITY_IMAGE=byconity/byconity:0.3.2
#BYCONITY_BINARY_PATH=~/ck/ByConity/build_docker/programs
BYCONITY_BINARY_PATH=

docker-compose -f docker-compose.essentials.yml -f docker-compose.simple.yml up -d

2. I executed the sql for several times.

[kevinthfang]

Did you deploy 'ClickHouse server 23.1.3' also using docker? about '16G Memery 8 Core 2.1Ghz CPU' is it restrained in docker config, or it's just your physical node resources?