Publishing site Mer 13 mar 2024 10:59:43 CET

blog/observability-for-microcks-at-scale/index.html

@@ -22519,7 +22519,7 @@
 
 
 
- <span class="text-muted fw-500"> 🗓️ March 12, 2024 <span class="mx-1 text-dark-50 fw-500">|</span> <i class="far fa-clock text-dark me-1"></i> 6 </span>
+ <span class="text-muted fw-500"> 🗓️ March 12, 2024 <span class="mx-1 text-dark-50 fw-500">|</span> <i class="far fa-clock text-dark me-1"></i> 8 </span>
 
  <h2 class="h3 my-3">Observability for Microcks at scale</h2>
 
@@ -23206,7 +23206,72 @@ <h2 id="walkthrough">Walkthrough</h2>
 
 >K6</a>
  which launches four different scenarios simultaneously during one minute. Here’s the raw output we got below with details on executed scenarios and raw performance metrics:</p>
-<p>And yes, we got this impressive <strong>756.5 hits/second</strong> with a p(90) response time of 28.2ms during the bench on a Macbook M2 with a 400MB heap! 🚀</p>
+<div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-shell" data-lang="shell"><span style="display:flex;"><span>$ docker run --rm -i -e BASE_URL<span style="color:#f92672">=</span><span style="color:#e6db74">${</span>MICROCKS_BASE_URL<span style="color:#e6db74">}</span> -e WAIT_TIME<span style="color:#f92672">=</span>0.1 grafana/k6:<span style="color:#e6db74">${</span>K6_VERSION<span style="color:#e6db74">}</span> run - &lt; bench-microcks.js
+</span></span><span style="display:flex;"><span>
+</span></span><span style="display:flex;"><span> /<span style="color:#ae81ff">\ </span> |‾‾| /‾‾/ /‾‾/ 
+</span></span><span style="display:flex;"><span> /<span style="color:#ae81ff">\ </span> / <span style="color:#ae81ff">\ </span> | |/ / / / 
+</span></span><span style="display:flex;"><span> / <span style="color:#ae81ff">\/</span> <span style="color:#ae81ff">\ </span> | <span style="color:#f92672">(</span> / ‾‾<span style="color:#ae81ff">\ </span>
+</span></span><span style="display:flex;"><span> / <span style="color:#ae81ff">\ </span> | |<span style="color:#ae81ff">\ \ </span>| <span style="color:#f92672">(</span>‾<span style="color:#f92672">)</span> | 
+</span></span><span style="display:flex;"><span> / __________ <span style="color:#ae81ff">\ </span>|__| <span style="color:#ae81ff">\_</span>_<span style="color:#ae81ff">\ \_</span>____/ .io
+</span></span><span style="display:flex;"><span>
+</span></span><span style="display:flex;"><span> execution: local
+</span></span><span style="display:flex;"><span> script: -
+</span></span><span style="display:flex;"><span> output: -
+</span></span><span style="display:flex;"><span>
+</span></span><span style="display:flex;"><span> scenarios: <span style="color:#f92672">(</span>100.00%<span style="color:#f92672">)</span> <span style="color:#ae81ff">4</span> scenarios, <span style="color:#ae81ff">85</span> max VUs, 2m45s max duration <span style="color:#f92672">(</span>incl. graceful stop<span style="color:#f92672">)</span>:
+</span></span><span style="display:flex;"><span> * browse: <span style="color:#ae81ff">20</span> looping VUs <span style="color:#66d9ef">for</span> 1m0s <span style="color:#f92672">(</span>exec: browse, gracefulStop: 30s<span style="color:#f92672">)</span>
+</span></span><span style="display:flex;"><span> * invokeRESTMocks: <span style="color:#ae81ff">200</span> iterations <span style="color:#66d9ef">for</span> each of <span style="color:#ae81ff">40</span> VUs <span style="color:#f92672">(</span>maxDuration: 2m0s, exec: invokeRESTMocks, startTime: 5s, gracefulStop: 30s<span style="color:#f92672">)</span>
+</span></span><span style="display:flex;"><span> * invokeGraphQLMocks: <span style="color:#ae81ff">100</span> iterations <span style="color:#66d9ef">for</span> each of <span style="color:#ae81ff">20</span> VUs <span style="color:#f92672">(</span>maxDuration: 2m0s, exec: invokeGraphQLMocks, startTime: 10s, gracefulStop: 30s<span style="color:#f92672">)</span>
+</span></span><span style="display:flex;"><span> * invokeSOAPMocks: <span style="color:#ae81ff">5</span> iterations <span style="color:#66d9ef">for</span> each of <span style="color:#ae81ff">5</span> VUs <span style="color:#f92672">(</span>maxDuration: 2m0s, exec: invokeSOAPMocks, startTime: 15s, gracefulStop: 30s<span style="color:#f92672">)</span>
+</span></span><span style="display:flex;"><span> 
+</span></span><span style="display:flex;"><span><span style="color:#f92672">[</span>...<span style="color:#f92672">]</span>
+</span></span><span style="display:flex;"><span>
+</span></span><span style="display:flex;"><span>running <span style="color:#f92672">(</span>1m04.0s<span style="color:#f92672">)</span>, 14/85 VUs, <span style="color:#ae81ff">10271</span> complete and <span style="color:#ae81ff">0</span> interrupted iterations
+</span></span><span style="display:flex;"><span>browse ↓ <span style="color:#f92672">[</span> 100% <span style="color:#f92672">]</span> <span style="color:#ae81ff">20</span> VUs 1m0s 
+</span></span><span style="display:flex;"><span>invokeRESTMocks ✓ <span style="color:#f92672">[</span> 100% <span style="color:#f92672">]</span> <span style="color:#ae81ff">40</span> VUs 0m12.7s/2m0s 8000/8000 iters, <span style="color:#ae81ff">200</span> per VU
+</span></span><span style="display:flex;"><span>invokeGraphQLMocks ✓ <span style="color:#f92672">[</span> 100% <span style="color:#f92672">]</span> <span style="color:#ae81ff">20</span> VUs 0m06.9s/2m0s 2000/2000 iters, <span style="color:#ae81ff">100</span> per VU
+</span></span><span style="display:flex;"><span>invokeSOAPMocks ✓ <span style="color:#f92672">[</span> 100% <span style="color:#f92672">]</span> <span style="color:#ae81ff">5</span> VUs 0m16.0s/2m0s 25/25 iters, <span style="color:#ae81ff">5</span> per VU
+</span></span><span style="display:flex;"><span>
+</span></span><span style="display:flex;"><span> ✓ status code should be <span style="color:#ae81ff">200</span>
+</span></span><span style="display:flex;"><span> ✓ pastryCall status is <span style="color:#ae81ff">200</span>
+</span></span><span style="display:flex;"><span> ✓ eclairCall status is <span style="color:#ae81ff">200</span>
+</span></span><span style="display:flex;"><span> ✓ eclairXmlCall status is <span style="color:#ae81ff">200</span>
+</span></span><span style="display:flex;"><span> ✓ eclairXmlCall response is Xml
+</span></span><span style="display:flex;"><span> ✓ millefeuilleCall status is <span style="color:#ae81ff">200</span>
+</span></span><span style="display:flex;"><span> ✓ allFilmsCall status is <span style="color:#ae81ff">200</span>
+</span></span><span style="display:flex;"><span> ✓ aFilmCall status is <span style="color:#ae81ff">200</span>
+</span></span><span style="display:flex;"><span> ✓ aFilmFragmentCall status is <span style="color:#ae81ff">200</span>
+</span></span><span style="display:flex;"><span> ✓ andrewCall status is <span style="color:#ae81ff">200</span>
+</span></span><span style="display:flex;"><span> ✓ karlaCall status is <span style="color:#ae81ff">200</span>
+</span></span><span style="display:flex;"><span> ✓ karlaCall body is correct
+</span></span><span style="display:flex;"><span> ✓ laurentCall status is <span style="color:#ae81ff">500</span>
+</span></span><span style="display:flex;"><span> ✓ laurentCall body is fault
+</span></span><span style="display:flex;"><span>
+</span></span><span style="display:flex;"><span> checks.........................: 100.00% ✓ <span style="color:#ae81ff">46385</span> ✗ <span style="color:#ae81ff">0</span> 
+</span></span><span style="display:flex;"><span> data_received..................: <span style="color:#ae81ff">132</span> MB 2.0 MB/s
+</span></span><span style="display:flex;"><span> data_sent......................: 7.8 MB <span style="color:#ae81ff">122</span> kB/s
+</span></span><span style="display:flex;"><span> http_req_blocked...............: avg<span style="color:#f92672">=</span>40.83µs min<span style="color:#f92672">=</span>291ns med<span style="color:#f92672">=</span>1.04µs max<span style="color:#f92672">=</span>18.4ms p<span style="color:#f92672">(</span>90<span style="color:#f92672">)=</span>5.04µs p<span style="color:#f92672">(</span>95<span style="color:#f92672">)=</span>8.37µs 
+</span></span><span style="display:flex;"><span> http_req_connecting............: avg<span style="color:#f92672">=</span>33.21µs min<span style="color:#f92672">=</span>0s med<span style="color:#f92672">=</span>0s max<span style="color:#f92672">=</span>18.35ms p<span style="color:#f92672">(</span>90<span style="color:#f92672">)=</span>0s p<span style="color:#f92672">(</span>95<span style="color:#f92672">)=</span>0s 
+</span></span><span style="display:flex;"><span> http_req_duration..............: avg<span style="color:#f92672">=</span>17.22ms min<span style="color:#f92672">=</span>1ms med<span style="color:#f92672">=</span>12.57ms max<span style="color:#f92672">=</span>782.36ms p<span style="color:#f92672">(</span>90<span style="color:#f92672">)=</span>28.2ms p<span style="color:#f92672">(</span>95<span style="color:#f92672">)=</span>36.67ms 
+</span></span><span style="display:flex;"><span> <span style="color:#f92672">{</span> expected_response:true <span style="color:#f92672">}</span>...: avg<span style="color:#f92672">=</span>17.21ms min<span style="color:#f92672">=</span>1ms med<span style="color:#f92672">=</span>12.57ms max<span style="color:#f92672">=</span>782.36ms p<span style="color:#f92672">(</span>90<span style="color:#f92672">)=</span>28.2ms p<span style="color:#f92672">(</span>95<span style="color:#f92672">)=</span>36.66ms 
+</span></span><span style="display:flex;"><span> http_req_failed................: 0.05% ✓ <span style="color:#ae81ff">26</span> ✗ <span style="color:#ae81ff">48709</span>
+</span></span><span style="display:flex;"><span> http_req_receiving.............: avg<span style="color:#f92672">=</span>80.43µs min<span style="color:#f92672">=</span>6.5µs med<span style="color:#f92672">=</span>22.66µs max<span style="color:#f92672">=</span>29.12ms p<span style="color:#f92672">(</span>90<span style="color:#f92672">)=</span>129.29µs p<span style="color:#f92672">(</span>95<span style="color:#f92672">)=</span>235.55µs
+</span></span><span style="display:flex;"><span> http_req_sending...............: avg<span style="color:#f92672">=</span>15.04µs min<span style="color:#f92672">=</span>1.58µs med<span style="color:#f92672">=</span>4.95µs max<span style="color:#f92672">=</span>9.27ms p<span style="color:#f92672">(</span>90<span style="color:#f92672">)=</span>22.33µs p<span style="color:#f92672">(</span>95<span style="color:#f92672">)=</span>36.83µs 
+</span></span><span style="display:flex;"><span> http_req_tls_handshaking.......: avg<span style="color:#f92672">=</span>0s min<span style="color:#f92672">=</span>0s med<span style="color:#f92672">=</span>0s max<span style="color:#f92672">=</span>0s p<span style="color:#f92672">(</span>90<span style="color:#f92672">)=</span>0s p<span style="color:#f92672">(</span>95<span style="color:#f92672">)=</span>0s 
+</span></span><span style="display:flex;"><span> http_req_waiting...............: avg<span style="color:#f92672">=</span>17.12ms min<span style="color:#f92672">=</span>902.5µs med<span style="color:#f92672">=</span>12.49ms max<span style="color:#f92672">=</span>782.34ms p<span style="color:#f92672">(</span>90<span style="color:#f92672">)=</span>28.08ms p<span style="color:#f92672">(</span>95<span style="color:#f92672">)=</span>36.49ms 
+</span></span><span style="display:flex;"><span> http_reqs......................: <span style="color:#ae81ff">48735</span> 756.508316/s
+</span></span><span style="display:flex;"><span> iteration_duration.............: avg<span style="color:#f92672">=</span>194.32ms min<span style="color:#f92672">=</span>10.5ms med<span style="color:#f92672">=</span>52.16ms max<span style="color:#f92672">=</span>5.57s p<span style="color:#f92672">(</span>90<span style="color:#f92672">)=</span>101.2ms p<span style="color:#f92672">(</span>95<span style="color:#f92672">)=</span>177.01ms
+</span></span><span style="display:flex;"><span> iterations.....................: <span style="color:#ae81ff">10285</span> 159.652981/s
+</span></span><span style="display:flex;"><span> vus............................: <span style="color:#ae81ff">14</span> min<span style="color:#f92672">=</span><span style="color:#ae81ff">14</span> max<span style="color:#f92672">=</span><span style="color:#ae81ff">85</span> 
+</span></span><span style="display:flex;"><span> vus_max........................: <span style="color:#ae81ff">85</span> min<span style="color:#f92672">=</span><span style="color:#ae81ff">85</span> max<span style="color:#f92672">=</span><span style="color:#ae81ff">85</span> 
+</span></span><span style="display:flex;"><span>
+</span></span><span style="display:flex;"><span>
+</span></span><span style="display:flex;"><span>running <span style="color:#f92672">(</span>1m04.4s<span style="color:#f92672">)</span>, 00/85 VUs, <span style="color:#ae81ff">10285</span> complete and <span style="color:#ae81ff">0</span> interrupted iterations
+</span></span><span style="display:flex;"><span>browse ✓ <span style="color:#f92672">[</span> 100% <span style="color:#f92672">]</span> <span style="color:#ae81ff">20</span> VUs 1m0s 
+</span></span><span style="display:flex;"><span>invokeRESTMocks ✓ <span style="color:#f92672">[</span> 100% <span style="color:#f92672">]</span> <span style="color:#ae81ff">40</span> VUs 0m12.7s/2m0s 8000/8000 iters, <span style="color:#ae81ff">200</span> per VU
+</span></span><span style="display:flex;"><span>invokeGraphQLMocks ✓ <span style="color:#f92672">[</span> 100% <span style="color:#f92672">]</span> <span style="color:#ae81ff">20</span> VUs 0m06.9s/2m0s 2000/2000 iters, <span style="color:#ae81ff">100</span> per VU
+</span></span><span style="display:flex;"><span>invokeSOAPMocks ✓ <span style="color:#f92672">[</span> 100% <span style="color:#f92672">]</span> <span style="color:#ae81ff">5</span> VUs 0m16.0s/2m0s 25/25 iters, <span style="color:#ae81ff">5</span> per VU
+</span></span></code></pre></div><p>And yes, we got this impressive <strong>756.5 hits/second</strong> with a p(90) response time of 28.2ms during the bench on a Macbook M2 with a 400MB heap! 🚀</p>
 <h2 id="conclusion">Conclusion</h2>
 <p>The Microcks user community expressed their wish to know what a single instance of Microcks is able to deliver in terms of throughput and response time. Through contribution we made <strong>Microcks ready to be fully observable</strong>, and we enabled optimization opportunities for very large-scale deployments. The project and the community users are now <strong>able to run benchmarks in autonomy to get figures and have a precise idea of what level of performance</strong> Microcks is capable of delivering.</p>
 <p>On a personal note, OpenTelemetry is the second-largest CNCF project and it&rsquo;s a big challenge to navigate in its ecosystem. This has been a good experience to find the nominal path to get the instrumentation, the storage of telemetry data, and visualization up and running for a real project. But, yes! I did it! 💪</p>