Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

Linux Foundation

Extending Container Performance Isolation - Regulating Memory Bandwidth and Cache in the Kernel

Linux Foundation via YouTube

Overview

Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn how to extend container performance isolation beyond CPU and memory capacity by regulating memory bandwidth and cache interference in the Linux kernel in this 42-minute conference talk from the Linux Foundation. Discover why containers currently provide limited protection against performance interference through shared CPU caches and memory bandwidth, which can increase application response times by 4-13x. Explore the limitations of the existing Linux resctrl infrastructure for controlling real-world applications, particularly its filesystem-based interface that lacks support for high-frequency measurement due to coarse-grained locking and filesystem overheads. Understand the memory interference problem and examine where current kernel mechanisms fall short in building controllers that can monitor and adjust quickly enough for application changes. Investigate proposed new kernel mechanisms designed to address these limitations, including a novel collector that captures per-process measurements of memory bandwidth usage at millisecond frequencies. Examine how this collector combines eBPF, high-resolution timers, and perf counters to achieve high-frequency monitoring, and learn how resctrl can be supported within this enhanced framework for better container performance isolation.

Syllabus

Extending Container Performance Isolation: Regulating Memory Bandwidth & Cache in... Jonathan Perry

Taught by

Linux Foundation

Reviews

Start your review of Extending Container Performance Isolation - Regulating Memory Bandwidth and Cache in the Kernel

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.