Hey,
I'm not sure if it's an issue or just something went wrong in my job configuration.
While using fio to benchmark s3 bandwidth with different block sizes, I got that with 500m+ block size I get more than 1gb/sec bw.
Trying to recreate these astonishing results using s3fs to read, I didn't even get close to fio results.
Is it an issue in fio, or something went wrong in my benchmarks?
These are the parameters served me in the fio job file:
[global]
ioengine=http
http_host=s3.us-east-1.amazonaws.com
http_mode=s3
http_s3_key=${S3_SECRET_KEY}
http_s3_keyid=${S3_KEY_ID}
http_s3_region=us-east-1
name=aws-test
filename=/mobileye-team-vd/idok/test_tv_parquet/20-03-31_17-08-46_Leo_Front_0005.snappy.parquet
rw=read
invalidate=1
randrepeat=0
direct=1
filesize=1300m
loops=10
numjobs=10
[test-1]
filename=/mobileye-team-vd/idok/test_tv_parquet/20-04-26_15-22-58_Leo_Front_0097.snappy.parquet
bs=1000m
filesize=1300m
Thanks ahead for any help or advice!
Sage
