Technical Guide: Optimizing ClickHouse Performance for Intel's High-Core-Count Processors
By
ashvardanian
An everything bagel for the brain. Substantive, layered, well-seasoned.
Summary
Intel engineers provide technical guidance on optimizing ClickHouse database performance for Intel's latest high-core-count processors, including Granite Rapids (128 P-cores per socket) and Sierra Forest (288 E-cores per socket). The article addresses the challenges and opportunities presented by the shift to "more, not faster cores" paradigm due to physical limitations, offering specific optimization strategies for analytical databases running on multi-socket systems with 400+ cores.
Key quotes
· 4 pulledIntel's latest processor generations are pushing the number of cores in a server to unprecedented levels - from 128 P-cores per socket in Granite Rapids to 288 E-cores per socket in Sierra Forest
These numbers multiply on multi-socket systems, such servers may consist of 400 and more cores
The paradigm of "more, not faster cores" is driven by physical limitations
For analytical databases like ClickHouse, ultra-high core counts represent a huge opportunity and a complex challenge
You might also wanna read
PostgreSQL Double Buffering: Why OS RAM and Buffer Cache Compete and How the 25% Rule Helps
This article explains the problem of double buffering in PostgreSQL, where the OS filesystem cache and PostgreSQL's shared_buffers both cach
Optimizing Top K Query Performance in PostgreSQL: Challenges and Solutions
This technical article examines the challenges of optimizing Top K queries in PostgreSQL databases, where 'Top K' refers to retrieving the K
paradedb.com·2mo ago
Optimizing Geospatial SQL Queries with H3 Indexes for 400× Performance Gains
The article explains how to dramatically speed up geospatial SQL queries (geo joins) using H3 indexes. It describes the performance problems
floedb.ai·3mo ago
Performance Optimization: Replacing Protobuf with Direct C-to-Rust Bindings in PgDog PostgreSQL Proxy
The article details how PgDog, a PostgreSQL proxy written in Rust, replaced Protobuf serialization with direct C-to-Rust bindings to achieve
pgdog.dev·4mo ago
Advanced PostgreSQL Optimization Techniques Beyond Conventional Methods
This article presents unconventional optimization techniques for PostgreSQL databases, moving beyond standard approaches like query rewritin
hakibenita.com·4mo ago
SQLite Query Optimization: Using Uncorrelated Subqueries to Skip Correlated Subqueries for Performance Gains
The article describes a SQLite query optimization technique where an uncorrelated scalar subquery is used to skip expensive correlated subqu
emschwartz.me·5mo ago