Online Shop Benchmarks

This folder contains 9 microservices from Googles Online Boutique ported as standalone serverless microbenchmarks.

Background

Online Boutique is a cloud-native microservices demo application. It consists of a 10-tier microservices application. The application is a web-based e-commerce app where users can browse items, add them to the cart, and purchase them.

For functionality details refer to the upstream repository (Online Boutique).

Porting to serverless kernel

We took the 9 microservices implementing the functionality of the application and integrated them into vSwarm. The current version of the sources is v0.3.9.

Note: To upgrade to another version see the update_remote_files.sh script.

Warning: This is just a script for your convenience. No guarantee that everything will work afterwards.

Functions summary

Benchmark	Knative infra	Tracing	Runtimes	Languages Implemented	gem5 support	Description
Cart Service	Serving	✕	docker, knative	C#	✕	Stores the items in the user's shopping cart in Redis and retrieves it.
Product Catalog Service	Serving	✕	docker, knative	Golang	✓	Provides the list of products from a JSON file and ability to search products and get individual products.
Currency Service	Serving	✕	docker, knative	Node.js	✓	Converts one money amount to another currency. Uses real values fetched from European Central Bank. It's the highest QPS service.
Payment Service	Serving	✕	docker, knative	Node.js	✓	Charges the given credit card info (mock) with the given amount and returns a transaction ID.
Shipping Service	Serving	✕	docker, knative	Golang	✓	Gives shipping cost estimates based on the shopping cart. Ships items to the given address (mock)
Email Service	Serving	✕	docker, knative	Python	✓	Sends users an order confirmation email (mock).
Checkout Service	Serving	✕	docker, knative	Golang	✕	Retrieves user cart, prepares order and orchestrates the payment, shipping and the email notification.
Recommendation Service	Serving	✕	docker, knative	Python	✓	Recommends other products based on what's given in the cart.
Ad Service	Serving	✕	docker, knative	Java	✕	Provides text ads based on given context words.

Running one of the benchmarks locally (using docker)

The detailed and general description how to run benchmarks local you can find here. The following steps show it on the adservice function.

Starting the function container

1. Build or pull the function images using make all-images or make pull.

Invoke once

2. Start the function with docker-compose

   docker-compose -f yamls/docker-compose/dc-adservice.yaml up

This will start the adservice function container together with the relay container which provides the common interface to vSwarm's benchmarking tools. The port of the function container is exposed to 9555 the relay can be reached via port 50000.

Invoking the function

With relay

In a new terminal, invoke the interface function with grpcurl.

# Same steps, we just invoke the relay at port 50000 instead of geo server directly
../../tools/bin/grpcurl -plaintext -import-path tools/invoker/proto/ -proto helloworld.proto  localhost:50000 helloworld.Greeter.SayHello

Invoke multiple times

Run the invoker with appropriate time and rps (requests per second)

# build the invoker binary
cd ../../tools/invoker
make invoker
# Specify the hostname through "endpoints.json"
echo '[ { "hostname": "localhost" } ]' > endpoints.json
# Start the invoker with a chosen RPS rate and time (With relay)
./invoker -port 50000 -dbg -time 10 -rps 1

Running this benchmark (using knative)

The detailed and general description how to run benchmarks on knative clusters you can find here. The following steps show it on the adservice function.

1. Build or pull the function images using make all-images or make pull.
2. Start the function with knative

   kn service apply -f ./knative_yamls/kn-adservice.yaml

3. Note the URL provided in the output. The part without the http:// we'll call $URL. Replace any instance of $URL in the code below with it.
4. Run the invoker

   # build the invoker binary
   cd ../../tools/invoker
   make invoker
   # Specify the hostname through "endpoints.json"
   echo '[ { "hostname": "$URL" } ]' > endpoints.json
   # Start the invoker with a chosen RPS rate and time
   ./invoker -port 80 -dbg -time 10 -rps 1

Tracing

We do not support tracing for this benchmark at the moment.