Enterprise Storage Basics: What is a SAN?

In this Basics Video, I review what a SAN is and why it's essential to Enterprise infrastructure.
Twitter: / intheshadaux
Reddit: / intheshadaux
Patreon: / intheshadaux

[@XxCrawdadCraigxX]

cool video

[@amlucent]

Let me start by saying it’s a good video and your grasp is good but you are confusing some terms.
A SAN is an acronym for the dedicated network to storage traffic. Storage Area Network. As you said this is done to segment and segregate the storage traffic from others. Optimize the traffic.
A lot of people mistakenly call the storage server (we usually just call it the array in the storage industry) serving block data a SAN. But technically that’s incorrect.
There are three over arching types of storage transmission categories.
1) Block - the storage server serves out an entire volume to the client. Sometimes called an extent or a LUN. The volume is raw - no file system. The client writes it’s own file system to the volume. In general, it’s a bad idea to share this volume between clients that are not in a “cluster” or cluster aware because it could cause both systems to read/write/alter the same files at the same time leading to corruption. Because it’s “block” or “raw” and the clients write their own file system it can be more performant and efficient than other types of storage.
Now you mentioned iscsi or fiber channel but that is just the transport method utilized to deliver the data. Iscsi is block over a regular TCP IP network (Ethernet). But there are others such as NVMEoCE or FCoE. I personally work a lot with fiber Channel because it is so much more efficient than tcpip for block storage traffic. Mostly I work with scsi-fc or NVMe-fc. This requires specialized fiber channel network cards and fiber channel switch infrastructure.
2) File - file is NAS. It’s a storage server ring it own file system with clients. There are lots of protocols historically speaking but the two main ones still in use today are SMB (often still called CIFS) and NFS. This is what you get when you browse a file share via network. It’s a great choice for sharing files and directories between computers that are not clustered or cluster aware since the storage server itself acts as the gatekeeper with the file system preventing the clients from accessing and writing over the same files at the same time (so prevents corruption).
3) Object - object is S3 storage or HTTP/S storage. You can interface with the files via gets, puts, and post commands. Really though object is a close cousin to NAS/FILE so much so that some enterprise storage vendors often lump it with file. You might have need for it in an application or in a container but right now youll need something like S3 browser to “browse” or enumerate it in some kind of way most people are familiar with.
So you’re using Truenas which does both Block and File, in the industry we call that a Unified Array.

[@InTheShadaux]

Love this comment! Adds a ton of value!
For those that are interested, I recommend looking up some of the terms he used. I do have plans to dive deeper into storage and I will for sure bring up your points.