README.md

PolyBench/C Microbenchmark

This experiment explores the baseline overheads of using WebAssembly instead of native x86 execution. We execute the PolyBench/C benchmark; a set of basic C kernels, and report the slowdown of WASM execution (using Granules) compared to native x86 execution.

Hardware Provisioning

First, provision the cluster. For ease of deployment, we still deploy a K8s cluster of just one node, which we will access directly.

inv cluster.provision --vm Standard_D8_v5 --nodes 1 --name ${CLUSTER_NAME}
inv cluster.credentials --name ${CLUSTER_NAME}

Native

Deploy the native baseline:

inv polybench.native.deploy

And run it:

inv polybench.run.native

To delete the k8s cluster run:

inv polybench.native.delete

Granny

First, deploy the K8s cluster:

cd ~/faasm
source ./bin/workon.sh
WASM_VM=wamr inv k8s.deploy --workers 1

Second, upload the WASM files:

inv polybench.wasm.upload

Third, run the experiments:

inv polybench.run.granny

To remove the cluster run:

cd ~/faasm
source ./bin/workon.sh
inv k8s.delete

Plotting the results

To plot the results you may run:

inv polybench.plot

which will generate a .pdf file in ./plots/polybench/slowdown.pdf.

Hardware Cleanup

Lastly, clean the cluster:

inv cluster.delete --name ${CLUSTER_NAME}