AI
Common GPU Operations and Fault Handling
· β˜• 18 min read
Content referenced or recorded while handling faults, continuously updated. 1. XID Error Events XID is NVIDIA’s error code. You can retrieve it with the command: 1 dmesg -T | grep -i "NVRM: Xid" or 1 journalctl --since `date -d "10 days ago" "+%Y-%m-%d"`|grep Xid The XID lets you locate the fault.

After a Month, I Finally Found an Approach to Writing an AI Agent
· β˜• 6 min read
1. Keep Trying to Ship the AI Application Side Based on my understanding of operations, I developed an open-source operations tool, https://github.com/shaowenchen/ops . The Ops tool divides operations work into two categories β€” script execution and file distribution β€” and both operations targets, hosts and Kubernetes clusters, implement both of

Creating a PVC Backed by JuiceFS (Enterprise Edition) in Kubernetes
· β˜• 4 min read
This post mainly records the script for creating an enterprise-edition JuiceFS PVC, to make configuration quick. 1. Set Environment Variables 1.1 Basic Configuration 1 2 3 export NAMESPACE= export PVC_NAME= export JUICEFS_VOLUME= 1.2 JuiceFS Credentials 1 2 3 export TOKEN= export ACCESS_KEY= export SECRET_KEY= 1.3 JuiceFS Service Configuration 1 2 export BASE_URL=http://x.

Creating a PVC Backed by JuiceFS (Community Edition) in Kubernetes
· β˜• 3 min read
This post mainly records the script for creating a community-edition JuiceFS PVC, to make configuration quick. For component deployment, see Managing Data in Kubernetes with Fluid and JuiceFS. 1. Set Environment Variables Bucket configuration 1 2 3 4 5 6 export ACCESS_KEY= export SECRET_KEY= export BUCKET= export ENDPOINT=ks3-cn-beijing-internal.ksyun.com export BUCKET_ENPOINT=$BUCKET.

Speeding Up Model Inference with TensorRT
· β˜• 4 min read
1. What Is TensorRT TensorRT is a C++ library mainly used for high-performance inference acceleration on NVIDIA GPUs. It provides both a C++ API and a Python API for integration. The mainstream deep learning frameworks TensorRT supports are: Caffe, where TensorRT can read the prototxt format directly TensorFlow, where the