![\"load-testing\"](\"https:\/\/devblogs.microsoft.com\/ise\/wp-content\/uploads\/sites\/55\/2024\/03\/load-testing.png\")
<\/p>\n<\/p>\n
After completing the initial phases of the API endpoints implementation, we needed to plan how we will test the performance, identify potential bottlenecks, and ensure that the system meets the Service Level Objectives (SLOs) requirements. Load testing is used for this purpose.<\/p>\n
Load testing is a type of performance testing that tests how the system behaves under varying load conditions, including the impact on responsiveness, throughput, and resource utilization.<\/p>\n
In a recent project, we evaluated different tools and decided to use K6<\/a> to conduct load testing. This post will go through the load testing concepts and shows how to use the K6 tool to perform load testing for API endpoints.<\/p>\n In this context, we have developed mock Customer API service using Spring Boot. This setup was used to illustrate the process of conducting load-testing with VS Code<\/a> and K6<\/a> which you can find in this repository<\/a>.<\/p>\n Why we need load testing:<\/strong><\/p>\n When you design your API tests, first you need to answer the following questions:<\/strong><\/p>\n The test goal determines the test type:<\/strong><\/p>\n The testing strategy likely follows something like this procedure:<\/strong><\/p>\n K6<\/strong> is an open-source load testing tool that makes performance testing easy and productive for engineering teams. It is Developer and Source control friendly and has a small learning curve(Javascript). You can run k6 locally, in the cloud, or as part of your CI\/CD pipeline.<\/p>\n Key Features at a Glance:<\/strong><\/p>\n K6 can be installed on Windows, Linux, and macOS. It can also be installed as a Docker image, or you can use the hosted version (k6 Cloud).<\/p>\n In this article, we will install K6 on Linux. For more information about installation on other platforms, please refer to the K6 documentation<\/a>.<\/p>\n K6’s test suite are written in JavaScript, making it easy to get started with performance testing. You can use the K6 CLI to run tests from the command line or integrate them with your automation scripts.<\/p>\n Each test scenario consists of three sections:<\/p>\n The More information about the test options can be found here<\/a><\/em>.<\/p>\n Following is an example of the Test Options section for the addCustomer endpoint test case.<\/p>\n The This section defines the actual testing logic, which could be one or more test scenarios and each scenario could test a single endpoint or a complete flow, It uses the K6 modules such as:<\/p>\n After creating the test script, we can execute it using the K6 CLI, which allows you to run single or multiple test files, and pass environment variables to the test execution.<\/p>\n The sample repository contains When the test is complete, k6 prints a top-level summary of the aggregated results to stdout. In addition to this default summary, k6 can output results in other formats including JSON, CSV, and more.<\/p>\n The summary report of the test script includes:<\/p>\n The follow image shows the test results for the addCustomer endpoint.<\/em><\/p>\n Every k6 test emits built-in metrics<\/a>. I will spotlight the most important HTTP metrics that can help measure the performance of the system and start to setup thresholds against the expected SLOs.<\/p>\n More information about metrics can be found here<\/a>.<\/em><\/p>\n In this post, we’ve demonstrated how to use the K6 framework to perform load testing for API endpoints, covering the selection criteria for choosing K6 as a load-testing tool as well as how to install K6 on Linux. We’ve shown how to easily build a load test case using JavaScript and how to execute the test case using the K6 CLI. We have also shared the test results and the different K6 metrics that can be used to measure the performance of the system.<\/p>\n K6 has proven to be a very powerful tool for load testing. It is easy to use and has a lot of features that can be used to build a comprehensive load test suite and can be seamlessly integrated into the development loop and CI\/CD pipelines.<\/p>\n In this post, We will go through the load testing concepts and how to use the K6 framework to perform load testing for API endpoints.<\/p>\n","protected":false},"author":119131,"featured_media":15351,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[1,3451],"tags":[3398,3472,3504,3503,3430,3346,3384],"class_list":["post-15350","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-cse","category-ise","tag-api","tag-http","tag-k6","tag-load","tag-performance","tag-testing","tag-vscode"],"acf":[],"blog_post_summary":" In this post, We will go through the load testing concepts and how to use the K6 framework to perform load testing for API endpoints.<\/p>\n","_links":{"self":[{"href":"https:\/\/devblogs.microsoft.com\/ise\/wp-json\/wp\/v2\/posts\/15350","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devblogs.microsoft.com\/ise\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devblogs.microsoft.com\/ise\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/ise\/wp-json\/wp\/v2\/users\/119131"}],"replies":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/ise\/wp-json\/wp\/v2\/comments?post=15350"}],"version-history":[{"count":0,"href":"https:\/\/devblogs.microsoft.com\/ise\/wp-json\/wp\/v2\/posts\/15350\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/ise\/wp-json\/wp\/v2\/media\/15351"}],"wp:attachment":[{"href":"https:\/\/devblogs.microsoft.com\/ise\/wp-json\/wp\/v2\/media?parent=15350"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/ise\/wp-json\/wp\/v2\/categories?post=15350"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/ise\/wp-json\/wp\/v2\/tags?post=15350"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}Load testing<\/h2>\n
\n
\n
\n
\n
K6<\/h2>\n
\n
K6 Installation<\/h2>\n
# Install K6 on Linux\r\n# https:\/\/grafana.com\/docs\/k6\/latest\/get-started\/installation\/#linux\r\nsudo gpg -k\r\nsudo gpg --no-default-keyring --keyring \/usr\/share\/keyrings\/k6-archive-keyring.gpg --keyserver hkp:\/\/keyserver.ubuntu.com:80 --recv-keys C5AD17C747E3415A3642D57D77C6C491D6AC1D69\r\necho \"deb [signed-by=\/usr\/share\/keyrings\/k6-archive-keyring.gpg] https:\/\/dl.k6.io\/deb stable main\" | sudo tee \/etc\/apt\/sources.list.d\/k6.list\r\nsudo apt-get update\r\nsudo apt-get install k6<\/code><\/pre>\nCreate test suite<\/h2>\n
\n
1. Test Options<\/h3>\n
Test Options<\/code> section defines test-run behavior, including the following:<\/p>\n\n
\n
\/**\r\n * Test configuration\r\n *\/\r\nexport const options = {\r\n tags: {\r\n test: 'customers-service',\r\n test_run_id: `Customers-Service-Load-${__ENV.K6_ENV}`,\r\n },\r\n thresholds: {\r\n \/\/ addCustomer thresholds\r\n 'http_req_failed{test_type:addCustomer}': ['rate<0.01'], \/\/ http errors should be less than 1%, availability\r\n 'http_req_duration{test_type:addCustomer}': ['p(95)<200'], \/\/ 95% of requests should be below 200ms, latency\r\n },\r\n scenarios: {\r\n \/\/ Load testing using K6 constant-rate scenario\r\n addCustomer_constant: {\r\n executor: 'constant-arrival-rate',\r\n rate: 10000, \/\/ number of iterations per time unit\r\n timeUnit: '1m', \/\/ iterations will be per minute\r\n duration: '1m', \/\/ total duration that the test will run for\r\n preAllocatedVUs: 2, \/\/ the size of the VU (i.e. worker) pool for this scenario\r\n maxVUs: 50, \/\/ if the preAllocatedVUs are not enough, we can initialize more\r\n tags: { test_type: 'addCustomer' }, \/\/ different extra metric tags for this scenario\r\n exec: 'addCustomer',\/\/ Test scenario function to call\r\n }\r\n }\r\n};<\/code><\/pre>\n2. Setup<\/h3>\n
Setup<\/code> section is optional and is used to set up the test environment, such as acquiring the authentication token and preparing the test initialization data.<\/p>\n\r\n\/**\r\n * prepare the test data like authentication\r\n * @returns Initial data for each test case\r\n *\/\r\nexport function setup() {\r\n \/\/ Get the authentication token\r\n const token = getAuthToken();\r\n return { access_token: token };\r\n}\r\n<\/code><\/pre>\n3. Test Scenario<\/h3>\n
\n
\/**\r\n * Add customer test case\r\n *\/\r\nexport function addCustomer(data) {\r\n\r\n const url = __ENV.CUSTOMERS_API_URL + \"\/\";\r\n const headers = {\r\n 'Authorization': `Bearer ${data.access_token}`,\r\n 'Content-Type': 'application\/json'\r\n };\r\n\r\n const payload = `{\r\n \"name\": \"Test Customer\",\r\n \"email\": \"email@test.com\"\r\n }`;\r\n\r\n var response = http.post(url, payload, { headers: headers });\r\n\r\n check(response, { 'status is 200': (r) => r.status === 200 });\r\n if (response.status != 200) {\r\n console.log(`operation: addCustomer, url: ${url}, Status:${response.status}`);\r\n }\r\n}<\/code><\/pre>\nExecute test script<\/h2>\n
# Execute load test using k6 CLI\r\nk6 run -e K6_ENV=local -e CUSTOMERS_API_URL=${CUSTOMERS_API_URL} customers-service.js<\/code><\/pre>\nmake commands<\/code> as shorthand to execute the test cases.<\/p>\n# Execute load test\r\nmake load-test<\/code><\/pre>\nTest Results<\/h2>\n
\n
![\"load-testing\"](\"https:\/\/devblogs.microsoft.com\/ise\/wp-content\/uploads\/sites\/55\/2024\/03\/load-testing-result.png\")
<\/p>\nK6 Results Metrics<\/h3>\n
\n
Conclusion<\/h2>\n
References<\/h2>\n
\n