1. Erasure Code Configuration The default erasure parity configuration is as follows: Erasure Set Size Default Parity (EC:M) 1 EC:0 2-3 EC:1 4-5 EC:2 6 - 7 EC:3 8 - 16 EC:4 The maximum parity is ERASURE_SET_SIZE/2, meaning the number of parity blocks cannot exceed the number of data blocks.

The earlier tests found the bottleneck in the NIC; this post adds supplementary testing under IPoIB. 1. MinIO Cluster Environment 1.1 Creating the MinIO Cluster 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 export CONTAINER_CLI=nerdctl export IMAGE=minio/minio:RELEASE.

1. Preparing the Configuration Files MinIO cluster 1 mc admin info local 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 ● minio1:9000 Uptime: 55 minutes Version: 2025-04-22T22:12:26Z Network: 2/2 OK Drives: 4/4 OK Pool: 1 ● minio2:9000 Uptime: 55 minutes Version: 2025-04-22T22:12:26Z Network: 2/2 OK Drives: 4/4 OK Pool: 1 ┌──────┬───────────────────────┬─────────────────────┬──────────────┐ │ Pool │ Drives Usage │ Erasure stripe size │ Erasure sets │ │ 1st │ 0.

VLLM Benchmark is a tool provided by VLLM for testing model performance, supporting a variety of inference backends. This article mainly records some of the process of using VLLM Benchmark to test model performance. 1. Starting the Model Service 1 2 3 4 5 6 7 8 9 10 11 12 13 14 python -m vllm.