Open Storage in the Enterprise with GlusterFS
summary
Built a production GlusterFS storage layer that serves three repeatable patterns and proves they work: a media service storing a petabyte with ~1 TB/day growth using Swift object access, bonded multi‑gigabit networks, RAID‑6 nodes and a load balancer for fast fingerprinting/transcoding; a self‑service pool that uses one big XFS per node with subdirectory bricks, per‑volume quotas and automation so only four Gluster commands provision tenants; and an active‑active, small‑file solution using Cassandra plus parallel geo-replication + Cassandra for active‑active small‑file replication (<30‑min SLA) that meets the customer SLA while improving small‑file performance; also exposes libgfapi with C/Python bindings and integrates Samba for multi‑protocol access.
AI-written summary · about
chapters (15)
- 0:00Start
- 3:03Heterogeneous Commodity Hardware
- 5:23What Is the Red Hat Storage
- 7:23Use Cases
- 9:39Implementation
- 11:45Multi Gigabit Network
- 16:35Quotas
- 21:36Performance Tune for Small File Worm Patterns
- 22:44Parallel Geo Replication
- 33:52Performance Requirements
- 35:04What Kind of Programmatic Interface Do You Have for Doing Reads and Writes to and from Luster
- 39:24And It Really Becomes a Limitation of Other Fairly Normal System Things How Many Processes Can I Run or How Much Memory Does each One of these Take Up When Do I Hit Limitations on My System for Most Practical Use Cases It's Just a Matter of Tuning for What You Expect To Use You Know Think Ahead of Time Are You Really Going To Run Hundreds or Thousands of Volumes and Bricks and if You Have a Case To Do that You Know Build the Systems and Tune Them Accordingly for that but Gloucester Itself Is Not Going To To Have It's a Real Limitation
- 39:46And if You Have a Case To Do that You Know Build the Systems and Tune Them Accordingly for that but Gloucester Itself Is Not Going To To Have It's a Real Limitation Do We Have Three Buttons There Is a Rebalance Operation That Happened So I Can Get a Little Technical Detail about the the Basic Hash Algorithm That Gloucester Does So It's Called the Distributed Hash Algorithm and It's It's Effectively Hashing on a Filename To Decide What Node a File Goes to So Every It's It the Decision It Makes on What Node To Put a File on Is Based on How Many Nodes Are in the System and What the File Name Is Right if
- 40:20What Node a File Goes to So Every It's It the Decision It Makes on What Node To Put a File on Is Based on How Many Nodes Are in the System and What the File Name Is Right if I Add an Additional Node Then It's Going To Change the the Decisions That the Distributed Hash Algorithm Would Make and Thus Put Files in a Different Location so that's the Long Answer To Effectively Yes if I Scale Out I Have To Run a Rebalance Operation every Time and the Rebalance Operation Has Really Two Activities That It Does One of Them Is To Basically Let the Distributed Hash Algorithm Know that the Architecture Is Changed so that New Files Can Get Written to the New Location but the Other the Other Piece of It Is Let's Take Existing Files
- 44:14Contact Information Feel Free To Take a Shot or You Can Go and Get the the Slides I Have I Have Not Uploaded Them Yet so that Url that People Got Redhead Com / D Black Will Be a Good Url I Have Not Put the Slides There Yet but I Will but My Contact Information As Far as Giving Information about Right at Storage or Gluster Communicating with Us on Social Media It's all There So if You Want on My Own Business Cards I'Ll I'Ll Dig Them out of My Backpack for You to any Other any Other Questions or Thoughts before We Wrap It Up Great Thanks Everybody for Coming
related talks
41mIntegration of flexible storage with the api of gluster
08mRapid Design and Effective Operating of a General Purpose Object Storage at RWTH
29mReliable and Cost-Effective: Open Storage Strategies for Kubernetes
31mScaling Autonomous-Driving R&D: A 100 PB, 100 Billion-File Open-Source Elastic File...
36mStorage Wars
29mTowards Sustainable Multi-Cluster Management in Real-World Academic Infrastructure
44mContainer Management at Scale
53mPrivate Cloud with OpenSource – "Now for Real!"