Scoring 2000 Products per Request: Performance Pitfalls in Golang
-
Ivan Sinitsin
Ozon Tech
Tech Lead
Data binning and understanding histograms
What this session covers
The session explains discretization by grouping continuous data into intervals (binning) in order to simplify the data, reduce noise, and improve algorithm performance. The value of binning lies in the fact that an application can efficiently predict user needs or avoid overloads. The value of the session is a few additional design tools for optimizing solutions operating on large data volumes.
World Congress 2026 Europe - Virtual Stage
Related talks
OLTP in the Lakehouse: Redefining Data for AI Workloads
-
Oleksandra Bovkun
Databricks
Developer Advocate
Challenges and Solutions for Efficient, Large-Scale Video Analysis
-
Magne Johansen
Norsvin
Senior System Developer
Systems Thinking for Performance: How to Diagnose and Fix Slow Systems without Guessing
-
Ivan Milanov
PubliCare
Engineering Manager