Skip to content

LPD-EPFL/ubft-artifacts

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

89 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Overview

uBFT is a BFT SMR system for microsecond applications.

This repository contains the artifacts and instructions needed to reproduce the experiments in our ASPLOS paper. More precisely, it contains:

  • Instructions on how to build the payloads for the different experiments.
  • Instructions on how to launch the experiments and obtain the results.

By running the experiments, you should be able to reproduce the numbers shown in:

  • Figure 7: End-to-end latency of different applications when either not replicated or replicated via Mu and uBFT’s fast path.
  • Figure 8: Median end-to-end latency for different request sizes of an unreplicated no-op application, as well as its latency when replicated with Mu, uBFT and MinBFT.
  • Figure 9: Decomposition of the latency of UBFT's fast and slow paths.
  • Figure 10: Median latency of multiple non-equivocation mechanisms for different message sizes.
  • Figure 11: uBFT's tail latency for different TCB tails t and different request sizes of 64B and 2KiB.
  • Table 2: uBFT replica memory usage for different TCB tails t and different request sizes.

Detailed instructions

This section will guide you on how to build, configure, and run all experiments, from scratch, in order to reproduce the results presented in our paper.

Cluster prerequisites

Running all experiments requires:

  • a cluster of 4 machines connected via an InfiniBand fabric,
  • Ubuntu 20.04 (different systems may work, but they have not been tested),
  • all machines to have FQDNs (further instructions on this matter will be given when needed),
  • all machines having the following ports open: 7000-7100, 11211, 18515, 9998
  • a machine with Intel SGX (could be the one with Inifiniband).

Deployment

The artifacts are built and packaged into binaries. Subsequently these binaries are deployed from a gateway machine (e.g., your laptop). The gateway machine requires the following depencencies installed to be able to execute the deployment scripts:

sudo apt install -y coreutils gawk python3 zip tmux

The cluster machines, assuming they are already setup for Infiniband+RDMA and Intel SGX, require the following dependencies to be able to execute the binaries:

sudo apt install -y coreutils gawk python3 zip tmux gcc numactl libmemcached-dev memcached redis

Generating the artifacts

Installing dependencies

The first step in generating the artifacts is building and packaging the binaries. To do so, you need the dependencies below.

Note: You can build and package the binaries in a cluster machine, the gateway or another machine. It is important, however, that you build the binaries in a machine with the same distro/version as the cluster's machines, otherwise the binaries may not work. For example, you can use a docker container to build and package the binaries. Alternatively, you can use one of the machines in the cluster.

Apt and PIP dependecies

Install the required dependencies on a vanilla Ubuntu 20.04 installation by running:

sudo apt-get -y install \
    python3 python3-pip \
    gawk build-essential cmake ninja-build \
    libmemcached-dev \
    libibverbs-dev # only if Mellanox OFED is not installed (see below).

pip3 install --upgrade "conan>=1.52.0"

Mellanox OFED dependency

Install the appropriate OFED driver by running:

wget http://www.mellanox.com/downloads/ofed/MLNX_OFED-5.3-1.0.0.1/MLNX_OFED_LINUX-5.3-1.0.0.1-ubuntu20.04-x86_64.tgz
tar xf MLNX_OFED_LINUX-5.3-1.0.0.1-ubuntu20.04-x86_64.tgz
sudo ./mlnxofedinstall

Intel SGX dependency

sudo echo 'deb [trusted=yes arch=amd64] https://download.01.org/intel-sgx/sgx_repo/ubuntu focal main' > /etc/apt/sources.list.d/intel-sgx.list
sudo apt-get -y install wget build-essential python pkg-config \
                                        libsgx-epid libsgx-quote-ex libsgx-dcap-ql \
                                        libsgx-enclave-common-dev libsgx-dcap-ql-dev libsgx-dcap-default-qpl-dev

cd /opt
sudo wget https://download.01.org/intel-sgx/sgx-dcap/1.12.1/linux/distro/ubuntu20.04-server/sgx_linux_x64_sdk_2.15.101.1.bin
sudo chmod +x /opt/sgx_linux_x64_sdk_2.15.101.1.bin 
sudo ./sgx_linux_x64_sdk_2.15.101.1.bin --prefix /opt
echo 'source /opt/sgxsdk/environment' | sudo tee -a /etc/profile
echo '/opt/sgxsdk/sdk_libs' | sudo tee /etc/ld.so.conf.d/sgx-sdk.conf
sudo ldconfig

Building the artifacts

Assuming all the machines in your cluster have the same configuration, you need to:

  • build all the necessary binaries, for example in a deployment machine,
  • package them and deploy them in all 4 machines.

The building process is long. Follow the instructions in each one of the following sub-directories:

  • Build Mu (on a deployment machine), as explained in mu-build.
  • Build uBFT (on a deployment machine), as explained in ubft-build.
  • Build MinBFT (on a deployment machine), as explained in minbft-build.
  • Build the Enclave code for MinBFT (on a deployment machine), as explained in usig-build.

Once you have finished with the above steps, the binaries will be packaged in payload.zip files.

Execute ./deployment.sh to generate the deployment.zip.

Deploying and running the tests

Transfer the deployment.zip to the gateway machine and extract its content.

unzip deployment.zip -d deployment
cd deployment

To run the experiments, follow the instructions under experiments.

Gateway: To run the experiments we assume the existence of gateway machine that has access to the 4-cluster machines and the sgx machine. This machine may be one of the deployment machines or e.g., your laptop. The gateway orchestrates the execution and gathers the experimental results.