Also known as: non-volatile memory
Volatile memory loses its contents the moment power is removed, while non-volatile memory keeps them — a distinction that shapes how every computer divides fast working memory from lasting storage.12
Overview
RAM is the archetypal volatile memory: it is fast and freely rewritable, which makes it ideal as the place a running program keeps its data, but everything in it vanishes at power-off. Dynamic RAM even needs constant refreshing to hold its charge while powered. Non-volatile memory — flash, ROM, hard disks, tape — gives up some speed or flexibility in exchange for retaining data without power.
Cache memory is also volatile, as are the CPU’s registers. The trade-off is fundamental: the fastest memories tend to be volatile, and persistence tends to cost speed or write endurance. That is why systems layer them rather than picking one.
Volatile vs non-volatile
The dividing line is simple, but it drives where each technology is used:
| Trait | Volatile | Non-volatile |
|---|---|---|
| Survives power-off | No | Yes |
| Examples | DRAM, SRAM, cache | Flash, ROM, disk, tape |
| Typical speed | Fastest | Slower |
| Typical role | Working memory | Persistent storage |
| Rewrites | Effectively unlimited | Often limited (flash wear) |
Where it fits
This split is why a computer cannot simply use one kind of memory for everything. Volatile RAM holds the working state of the operating system and applications; non-volatile storage holds files and programs between sessions, which is why work must be saved to disk to survive a reboot. The memory hierarchy arranges these layers by speed and cost. For GopherTrunk, decoded frames live briefly in volatile RAM as they are processed, then are written to non-volatile storage so the log survives a restart or power cut at the antenna site.
Sources
-
Volatile memory — Wikipedia, on memory that requires power to retain data. ↩
-
Non-volatile memory — Wikipedia, on memory that retains data without power. ↩