Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Performance testing: You need to think critically

Avatar for Roman Ananyev Roman Ananyev
February 03, 2020

Performance testing: You need to think critically

How to measure the performance of the CS-Cart project simply and without technical knowledge :) And of course, why all the online-performance-testing services lie, or dont provides a clear view ;)
https://www.simtechdev.com/cloud-hosting

---
Cloud hosting for CS-Cart, Multi-Vendor, WordPress, and Magento
by Simtech Development - AWS and CS-Cart certified hosting provider
free installation & migration | free 24/7 server monitoring | free daily backups | free SSL | and more...

Avatar for Roman Ananyev

Roman Ananyev

February 03, 2020

More Decks by Roman Ananyev

Other Decks in Technology

Transcript

  1. Why performance tests suck sometimes? 1. Slow or out-of-date device

    or browser (laptop, mobile phone, ...) 2. Slow network speed / Low throughput network, some latency issues 3. Any tests without metrics (numbers) a. It looks that … b. I think ...
  2. When performance tests suck? 1. You use third-party services and

    do not read or analyze their reports a. You are scared by the score but you don’t understand what it means b. You don't understand the parameters the test measures 2. You don’t know how, why and what you should test a. And what to do with the results
  3. Data, Analysis, Metrics, Knowledge: This is the key • All

    services will show you only that they want to ◦ and won’t provide the clear view ◦ they want to sell you something (extra checks, bigger reports, ...) ◦ … • You don’t know if the score is correct ◦ Mobile networks from the out-of-date devices ◦ Old servers with HDD disks ◦ …
  4. How-to: Find CS-Cart/MV performance issues 1. Log in to CS-Cart/MV

    admin panel and add to URL “?debug=true” ◦ Should be like https://my-great-store.com/adminpanel.php?debug=true 2. Open the slow page of the website 3. Click the “bug” ◦ Page generating time 0.0483 s (bottom right corner) i. If more than 0.5 s → issues in the code and SQL queries ◦ SQL: 218 queries 0.0668 s (right sidebar) i. If more than 500 requests → issues in the code ii. If takes more than 0.2-0.3 s → issues in the code and SQL queries ◦ Blocks: 20 (0 from cache) blocks rendered in 0.2736 s (right sidebar) i. If 0-5 blocks from the cache → code side issues (broken caching system) ii. If rendering takes more than 0.5 s → issues in the code and SQL queries https://www.simtechdev.com/cloud/wiki/user/find-cscart-mv-performance-issues/
  5. Can we provide a performance report? Sure, we can do

    it for you with some extra features • ✅ Website PHP/SQL profiling report with detection of bottlenecks • ✅ Load-testing • ✅ Reports about PHP errors and notices • ✅ Report about issues in the code ◦ ✅ Based on Best Practices for each CMS and PHP Standard Recommendation (PSR) ✅ Included in our Cloud Hosting Solution
  6. You need to think critically Google pagespeed report 1. amazon.com:

    75 98 2. bestbuy.com: 61 96 3. etsy.com: 59 94 4. aliexpress.com: 46 65 5. ebay.com: 45 75 6. walmart.com: 41 91 7. asos.com: 37 92 8. apple.com: 34 88 9. wish.com: 2 24 (Wish is primarily a mobile marketplace )
  7. Conclusion 1. TTFB is a complex metric a. It's mostly

    a metric of the application and the quality of its code b. It’s a metric of PHP and SQL queries 2. Performance testing should be done manually with result analysis a. All third-party performance testing systems don't give a damn about quality b. All these testing systems have their own “premium” checkers and optimizations i. They want to sell it in the first place, not to help you 3. A lot of extra things make your website slower a. Old and out-of-date devices (PC, modems or routers, laptops, mobile phones, ...) b. Old software (OS, browsers, drivers, ...) c. Slow speed or limited access networks, … d. …
  8. Don't hesitate to contact me --- Sincerely yours, Roman Ananev,

    Head of SRE & Cloud Hosting cloud (at) simtechdev.com rananev (at) simtechdev.com linkedin.com/in/rananev
  9. simtechdev.com: is this metric correct? (: pǝssǝɔɔɐ ǝq ʇuɐɔ suıɐɯop

    ǝɹɐ ǝɹǝɥʇ ǝsnɐɔǝq ‘ʇɔǝɾoɹd ǝɥʇ oʇ ǝlqɐɔılddɐ ʇou ˙ʇɔǝɹɹoɔuı
  10. simtechdev.com: is this metric correct? (: pǝssǝɔɔɐ ǝq ʇuɐɔ suıɐɯop

    ǝɹɐ ǝɹǝɥʇ ǝsnɐɔǝq ‘ʇɔǝɾoɹd ǝɥʇ oʇ ǝlqɐɔılddɐ ʇou ˙ʇɔǝɹɹoɔuı
  11. simtechdev.com: is this metric correct? (: pǝssǝɔɔɐ ǝq ʇuɐɔ suıɐɯop

    ǝɹɐ ǝɹǝɥʇ ǝsnɐɔǝq ‘ʇɔǝɾoɹd ǝɥʇ oʇ ǝlqɐɔılddɐ ʇou ˙ʇɔǝɹɹoɔuı
  12. simtechdev.com: is this metric correct? (: pǝssǝɔɔɐ ǝq ʇuɐɔ suıɐɯop

    ǝɹɐ ǝɹǝɥʇ ǝsnɐɔǝq ‘ʇɔǝɾoɹd ǝɥʇ oʇ ǝlqɐɔılddɐ ʇou ˙ʇɔǝɹɹoɔuı
  13. simtechdev.com: is this metric correct? (: pǝssǝɔɔɐ ǝq ʇuɐɔ suıɐɯop

    ǝɹɐ ǝɹǝɥʇ ǝsnɐɔǝq ‘ʇɔǝɾoɹd ǝɥʇ oʇ ǝlqɐɔılddɐ ʇou ˙ʇɔǝɹɹoɔuı
  14. Any questions? --- Sincerely yours, Roman Ananev, Head of SRE

    & Cloud Hosting cloud (at) simtechdev.com rananev (at) simtechdev.com linkedin.com/in/rananev