Skip to content

E-commerce search benchmark is the first end-to-end application benchmark for e-commerce search system with personalized recommendations.This work is joint with Prof. Jianfeng Zhan (http://www.benchcouncil.org/zjf.html) 's team, who is also the chair of International Open Benchmark Council (BenchCouncil, http://www.benchcouncil.org/).

License

Notifications You must be signed in to change notification settings

alibaba/eCommerceSearchBench

Repository files navigation

e-commerce search benchmark

Introduction

E-commerce search benchmark is the first end-to-end application benchmark for e-commerce search system with personalized recommendations. It helps people deeply understand the characteristics of workloads for e-commerce search and drives better design options for industry search system.It specifies an e-commerce search dataset and workloads that driven by production data and real-world user queries, and targets at hardware and software systems that provide e-commerce online search service. Therefore, any of those system can be used to establish the feasibility of this benchmark.

Features of the benchmark:

  • Provides a Data Generator using real-world datasets and producing synthetic data of various scales.
  • Provides a Workload Generator that driven by the real-world user logs from Taobao.
  • Provides a e-commerce search model eSearchEngineModel that simulated the search system of Taobao.
  • Evaluates the overall performance and performance of individual components. arch

The e-commerce search benchmark is built on docker images. As shown in the figure above,the benchmark consists of 7 docker images:

  1. aliesearch-search-planner
  2. aliesearch-query-planner
  3. aliesearch-tf-serving
  4. aliesearch-ha3
  5. aliesearch-ranking-service
  6. aliesearch-jmeter-image
  7. aliesearch-benchmark-cli

where, images 1~5 constitute the e-commerce search model eSearchEngineModel, aliesearch-benchmark-cli is the Data Generator, aliesearch-jmeter-image drives the workload generated by Workload Generator to the e-commerce search model eSearchEngineModel.

Preparation

Dependency

The following build tools are required:

  • gradle 4.x
  • maven 3.x
  • docker 17.09+
  • jdk8

Note: Make sure you can use docker without sudo by running

sudo usermod -aG docker $USER

Build

Run build.sh in eSearchEngineModel directory to build and publish image, etc.

  • Compile and build docker images

    ./build.sh build
  • Publish the docker images to specified remote docker reposition. If you are only running locally, you can skip this step.

    ./build.sh push ${repo_name}

Running benchmark

Note:

Correctly value for java maximum memory size must be required, otherwise the docker images won't work. You can set it for temporary by running

sudo sysctl -w vm.max_map_count=262144

or you can set it for permanent by directly edit the /etc/sysctl.conf file on the host, adding a line as flollow:

vm.max_map_count = 262144

and then run

sysctl -p

Running an Example in standalone

  1. Copy appctl.sh in this directory to your working directory;
  2. Pull all the images by running: (If you are only running locally, do skip this step.)
     sudo ./appctl.sh pull
    
  3. Start all images and automatically import the default data by running:
     sudo ./appctl.sh start
    
  4. Check if all the images are ready to work by running the following command in any docker:
     sudo docker exec -it aliesearch-jmeter-image bash
     curl -H 'Content-Type:application/json;charset=UTF-8' -d'
                 {"uid":"798", "page":0, "query":"68"}' ${search_planner_ip}:8080/search
    
  5. Run an example experiment by jmeter:
    • login to 'jmeter-image' by running:
    sudo docker exec -it aliesearch-jmeter-image bash
    
    • run to start the pressure test process by running:
    cd apache-jmeter-5.1.1
    ./bin/jmeter -n -t search_stress.jmx -l result -e -o report
    

Running an Example on k8s

Distributed deployment is also provided depending on k8s, which is more in line with the online environment. The benchmark can be run on a k8s cluster as follows:

  1. Copy k8s.yml to working directory on k8s master node.
  2. Replace the docker images repo configuration ink8s.yml with your own repo
  3. Pull all images and automatically import the default data by running:
    sudo kubectl create -f k8s.yml
    
  4. Check the pods status to ensure that all pods start up properly by running:
    sudo kubectl get pod
    
  5. Run an example experiment by jmeter:
    • log in to jmeter pod
      sudo kubectl exec -it aliesearch-jmeter-image bash
      
    • run to start the pressure test process
      cd apache-jmeter-5.1.1
      ./bin/jmeter -n -t search_stress.jmx -l result -e -o report
      

Running with Customized Pattern

The benchmark comes with an e-commerce data generator and workloads generator that driven by production data and real-world user queries. There are 4 steps for running an experiment with Customized data scale and workload pattern.

1. Install the benchmark

Start all the images of the benchmark according to the 1~3 steps in Running Example in standalone

2. Data Generation

Login to the benchmark-cli image by running:

sudo docker exec -ti aliesearch-benchmark-cli bash

Generate goods and user data , load them into corresponding search components by running:

vim entrypoint.sh
sh entrypoint.sh ${scale_factor}

where, ${scale_factor} sets the scale factor determining the dataset size (1 scale factor equals 10K goods and 6K user, 10 scale factor equals 100K goods and 60K user,and so on)

3. Workload Generation

Change to the directory workload_generator, and generate workload for the specified start hour of a day (-t) , the duration of the workload (-d), the workload factor to genearte (-f), and the user scale (-u) , by running:

python3 workload_generator.py -t 21 -d 1800 -f 0.1 -u 100000

A workload_u100000_h21_d1800_f0.1.csv file will be generated in the directory workload_generator, which will be driven to system model under test (e.g. eSearchEngineModel ).

4. Run experiment by jmeter

Firstly, Copy the generated workload file query_workload.csv to jmeter path (apache-jmeter-5.1.1/bin/jmeter) in jmeter docker. Then, login to 'jmeter-image' by running:

sudo docker exec -ti aliesearch-jmeter-image bash

And change to the jmeter bin directory, start the pressure test process by running:

cd apache-jmeter-5.1.1
./bin/jmeter -n -t search_stress.jmx -l result -e -o report

Running Benchmark in batches

  1. Change directory to ./run-scripts;
  2. Pull or build all the images
  3. Start all images with default dataset
    • for running Benchmark in standalone, running:
    ./deploy_standalone.sh start
    
    • for running Benchmark in a cluster, running:
    ./deploy_cluster.sh start
     with specified ip address for each docker in the `multi_iplist_env.sh` file
    
  4. Check if all the images are ready to work by running:
     curl -H 'Content-Type:application/json;charset=UTF-8' -d'
                 {"uid":"798", "page":0, "query":"68"}' ${search_planner_ip}:8080/search
    
  5. Run experiments in batches by running the following script with specified cases:
    ./run_batch_cluster.sh
    
  6. Analyze the results which are gathered in ./run-scripts/jmeter_result:
    • Change directory to ./run-scripts/jmeter_result and Preprocess the result logs by running:
      ./result_stat.sh
      
    • The results consist of QPS, response time, latency breakdown of response time , system metrics, and so on.

Running Benchmark in arm64v8 based platform

The build.sh in eSearchEngineModel directory supports to build images for arm64v8 based platform when running on arm64v8 based platform. And all the experimennts can be run on arm64v8 based platform following the steps descripted above.

About

E-commerce search benchmark is the first end-to-end application benchmark for e-commerce search system with personalized recommendations.This work is joint with Prof. Jianfeng Zhan (http://www.benchcouncil.org/zjf.html) 's team, who is also the chair of International Open Benchmark Council (BenchCouncil, http://www.benchcouncil.org/).

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published