Building a RESTful real-time analytics system with Pyramid

on YouTube ↗26m

" summary

CeleraOne describes a production system: an in‑memory C++ engine plus a REST layer built with Pyramid running under uWSGI, a proxy layer using OpenResty/LUA, Django for internal tooling, and a Django/Jenkins packaging pipeline. Workers aggregate and cache precomputed metrics into MongoDB; full clusters serve up to 10,000 requests per second. The claim is that this combination scales to billions of impressions and real‑time user segmentation; deployment and tests run end‑to‑end and are reported as working.

AI-written summary · about

" chapters (9)

1. 0:00Intro
2. 1:47C1 platform architecture
3. 5:10Typical setup
4. 6:57Python Software stack
5. 8:32Pyramid app with Cornice and Colander
6. 12:24Test case example
7. 13:39Pyramid app - modularity and pluggability
8. 15:36Pyramid app - pluggability example
9. 21:21Testing, packaging, deployment

" related talks

• 27mPython, Pyramid, and Embedded DevicesJustin Clacherty · PyCon Australia 2014 Science & Data miniconf
• 47mBuilding Scalable Web AppsDavid Cramer · EuroPython 2011
• 19mOptimizing Data Retrieval with Python CeleryJenna Conn, Hannah Cline · PyCon US 2021
• 41mEmbracing the Monolith in Small TeamsLeon Sasson · PyData Chicago 2016
• 26mBuilding A Python Service Using Open Source Components That Connect, Transform & Deliver Data In RTTomas Neubauer · Big Data LDN 2022
• 40mPython and pandas as back end to real-time data driven applicationsGiovanni Lanzani · PyData Berlin 2014
• 21mHigh-Availability DjangoFrankie Dintino · DjangoCon US 2016
• 42mBuilding a hosting platform with PythonAndrew Godwin · EuroPython 2011