Benedikt Riedel
Benedikt Riedel
ClusterRoles for Pulsar Functions, Promotheus, etc. This is related to issue #173 `rbac.limit_to_namespace: true` only limits required pulsar pieces (broker, bookie, etc.) to the namespace, but not Pulsar function workers,...
When inputting a single input array with ``` inputs = [ httpclient.InferInput("input_branch1", self.model_input_shape, "FP32") ] outputs = [ httpclient.InferRequestedOutput("Target1") ] inputs[0].set_data_from_numpy(input_data[0].astype(np.single)) ``` I get the following error: `tritonclient.utils.InferenceServerException: [400] [request...
Need to change the relative dir in terraform_init.sh -> terraform -chdir=../../contrib/terraform/openstack init terraform_apply.sh -> terraform -chdir=../../contrib/terraform/openstack apply -auto-approve -var-file=/home/briedel/code/ewms/jetstream_kubespray/inventory/test_cluster/cluster.tfvars -var="cluster_name=$CLUSTER" -var="network_name=${CLUSTER}-network" /jetstream_kubespray/contrib/terraform/openstack/modules/compute/main.tf -> Absolute dirs needed
We are running `tritonclient[http]=2.41.0` with server running `24.06-py3`. When there are O(600) requests reaching the server we intermittently receive the following error from triton: ``` Traceback (most recent call last):...
**Description** Triton does not clear or release GPU memory when there is a pause in inference. In the attached diagrams the same model is being used. It is served via...
The quick start guide for IceCube in cvmfs [1] suggests using-v4.2.1. We have an updated version (py3-v.4.30) available and a new version (py3-v4.4.0) will be coming out this month (Jan...
I am trying to use the PyTorch GPUU devcontainer. I am running into a weird issue with `/tmp` not being writable. This may be related to #1480 and https://github.com/devcontainers/images/issues/1556. The...