storage
storage copied to clipboard
MLPerf™ Storage Benchmark Suite
Hi, I have some questions. 1. I want to run the benchmark on an nvme disk that is not mounted on the file system. However, it seems that this evaluation...
hello team, I install depends follow the guide, but I suffer this on FedoraOS37 ``` (base) [root@cassiopeia storage]# pip3 install -r dlio_benchmark/requirements.txt Looking in indexes: https://pypi.org/simple, https://download.pytorch.org/whl/cpu, https://developer.download.nvidia.com/compute/redist Collecting absl-py==1.3.0...
We install the v1.0 version on an ARM server. For dlio-profiler, there is no aarch64 version of whl. It takes us a lot of effort to compile this packpage from...
Hi, The benchmark failed with the following error when running on Kubernetes. I was able to workaround it by setting environment variable `RDMAV_FORK_SAFE=0`, but not sure whether there is any...
How do I apply MLPerf Storage benchmark to Ceph using the S3 API
While executing the benchmark.sh script to run the benchmark, I was getting the following error: ./benchmark.sh run --workload unet3d --num-accelerators 1 --results-dir unet3d_results --param dataset.data_folder=/raid/unet3d_data/ --param dataset.num_files_train=37500 [INFO] 2023-08-01T14:57:33.254442 Running...
I tried to run the given command: ./benchmark.sh run --workload unet3d --num-accelerators 4 --results-dir unet3d_results --param dataset.data_folder=unet3d_data But got this: Error executing job with overrides: ['workload=unet3d', '++workload.workflow.generate_data=False', '++workload.workflow.train=True', '++workload.dataset.data_folder=unet3d_data', '++workload.workflow.profiling=False',...
Hi there, This is similar to https://github.com/mlcommons/storage/issues/22. I'm following all the instructions [here](https://github.com/mlcommons/storage#u-net3d) but still received an error about the directory structure. The full error message with the command is...
I'm working on a project that needs to run a profiler with the benchmark. I see that ++workload.workflow.profiling and ++workload.profiling.profiler are in EXTRA_PARAMS and disabled. Would it be possible to...
In the storage WG meeting, some people were looking for a way to use container to run the benchmark.