Longhorn: Intro, Deep Dive and Q&A

on YouTube ↗32m

" summary

Longhorn ships a new v2 data engine based on SPDK that shows clear IOPS and throughput gains over v1 in CNF-sponsored benchmarks, scaling with 1–4 CPU cores per volume; v2 supports MV over fabric and UBRock frontends and aims to run on low- to high-spec nodes. Upcoming 1.8 adds snapshot checksum online replica rebuilding, volume auto-salvage, live migration and live upgrade and multiple backup targets; the project reports ~159,000 nodes running Longhorn and a predictable Jan/May/Sep release cadence.

AI-written summary · about

" chapters (11)

1. 0:00Longhorn intro and mission
2. 2:10What is Longhorn
3. 4:20Project growth and adoption
4. 7:10Longhorn architecture
5. 13:14Longhorn data engines v1 & v2
6. 18:13Volume modes and CSI
7. 20:45Backup compression and protection
8. 24:41Longhorn v2 performance
9. 28:38Longhorn v2 benchmarks
10. 30:25Community highlights
11. 31:51Q&A

" related talks

• 31mLonghorn: Intro, Deep Dive and Q&ADerek Su · KubeCon + CloudNativeCon Japan 2026
• 32mLonghorn: Intro, Deep Dive and Q&AJoshua Moody, Sheng Yang · KubeCon + CloudNativeCon Europe 2022
• 38mLonghorn: Intro, Deep Dive And Q+ADavid Ko, Joshua Moody · KubeCon + CloudNativeCon North America 2022
• 35mImproving Longhorn Performance With SPDKDavid Ko, Keith Lucas · KubeCon + CloudNativeCon North America 2022
• 47mLonghorn: Intro, Deep Dive and Q+ADavid Ko, Shuo Wu · KubeCon + CloudNativeCon + Open Source Summit China 2023
• 48mIntro to Longhorn: Open Source Cloud-Native Distributed Block Storage Built On and For K8sSheng Yang · KubeCon + CloudNativeCon North America 2019
• 37mBuild a Kubernetes Based Cloud Native Storage Solution From ScratchSheng Yang · KubeCon + CloudNativeCon Europe 2019
• 24mSheng Liang, Rancher Labs & Murli Thirumale, Portworx | KubeCon + CloudNativeCon Europe - VirtualSheng Liang, Murli Thirumale · KubeCon + CloudNativeCon Europe 2020