Flow Classification Optimizations in DPDK

preview_player
Показать описание
This presentation will cover flow table design using DPDK, as well as new algorithmic and hardware optimizations for the  RTE hash library that improves on the lookup and the flow update/insert performance. Furthermore, a new research proof of concept (POC) to optimize the OVS flow lookup using a two-layer lookup technology based on DPDK libraries will be highlighted with some preliminary research results.
  1. DPDK Project
Рекомендации по теме
Flow Classification Optimizations 0:43:04

Flow Classification Optimizations in DPDK

OvS-CD: Optimizing Flow 0:20:07

OvS-CD: Optimizing Flow Classification for OvS using the DPDK Membership Library - Sameh Gobriel

Extending DPDK Flow 0:31:34

Extending DPDK Flow Classification Libraries for Determinism and Cloud Users

Partial offload optimization 0:07:46

Partial offload optimization and performance on Intel Fortville NICs using rte flow

Flow-based Packet Processing 0:31:25

Flow-based Packet Processing Framework on DPDK and VPP - Hongjun Ni, Intel

3.3: DPDK Optimizations 0:59:18

3.3: DPDK Optimizations

DPDK Flow Classification 0:15:37

DPDK Flow Classification and Traffic Profiling & Measurement - Ren Wang & Yipeng Wang, Inte...

OvS DPDK Acceleration 0:07:19

OvS DPDK Acceleration with rte flow Challenges and Gaps

OvS-DPDK Performance Optimizations 0:27:17

OvS-DPDK Performance Optimizations to Meet Telco Needs by Jan Scheurich and Mark Gray

Rte_flow Optimization in 0:26:35

Rte_flow Optimization in i40e Driver - Chenmin Sun, Intel

Classification optimizations to 0:30:19

Classification optimizations to enable megaflow offload onto lightweight HW pipelines

Hierarchical Flow Classification 0:17:22

Hierarchical Flow Classification Enhancements for OVS MegaFlow Cache by Sameh Gobriel

Open vSwitch DPDK 0:17:27

Open vSwitch DPDK Acceleration Using HW Classification

Flow Offloads for 0:30:43

Flow Offloads for DPDK Applications: The Partial, The Full, and The Graceful - Mesut Ali Ergin

Implementation of Flow-Based 0:09:36

Implementation of Flow-Based QoS Mechanism with OVS and DPDK - Karuppusamy M

OVS-DPDK Performance Benchmark 0:07:25

OVS-DPDK Performance Benchmark and Analysis with Multi-VMs

Tungsten Fabric Performance 0:12:15

Tungsten Fabric Performance Optimization by DPDK - Lei Yao, Intel

The Discrepancy of 0:08:03

The Discrepancy of the Megaflow Cache in OVS, Final Episode

Introduction to DPDK 0:11:38

Introduction to DPDK

DPDK Memory Allocation 0:36:35

DPDK Memory Allocation Optimization - Liang Ma, TikTok

Using nDPI over 0:30:46

Using nDPI over DPDK to Classify and Block Unwanted Network Traffic - Luca Deri, ntop

2.7: Data-Plane Software 0:29:17

2.7: Data-Plane Software Optimization

Can OVS-DPDK be 0:20:02

Can OVS-DPDK be further optimised?

[ENG] Marian Marinov: 0:40:48

[ENG] Marian Marinov: Comparison of eBPF, XDP and DPDK for packet inspection / #LinuxPiter

Комментарии
Автор

How do you plan to fill your user space RX/TX queues? PF_RING or PACKET_MMAP with an AF_PACKET socket?
What other options do you have in bypassing the kernel?
Why not build your abstraction on top of DPDK (you get HW offload abstractions for crypto, compression for free as well)?

lankasiri