Dynamic random-access memory
Each bit is a charge on a tiny capacitor, refreshed many times a second.
Computing and AI · Data storage · No. 283
- Electronic age
- Readiness 9 of 9
- Component
Your computer's main memory forgets everything within moments unless each bit is topped up several times a second.
Chart: Illustration: Dynamic random-access memory
Where it comes from
Chart: Map of where it comes from
After an IBM review in the autumn of 1966, Robert Dennard wondered whether a small capacitor could replace the magnetic ring of core memory. By 1967 he had reduced a memory cell to one transistor and one capacitor.
Three things to know
- A DRAM cell is just one transistor and one capacitor: a charged capacitor means 1 and an empty one means 0.
- Intel's 1103 of 1970 held 1024 bits at 1 cent per bit and quickly pushed magnetic core memory out of computers.
- Memory chips were the second-largest chip category in 2024, with sales of 165.1 billion US dollars, up 78.9%.
The journey
- 1 Silicon waferVery pure silicon, as for any chip
- 2 CellEach bit gets one MOS transistor and one tiny capacitor
- 3 ArrayBillions of cells are laid out in rows and columns
- 4 Chip and moduleChips are packaged and mounted on memory modules
- 5 UseThe computer reads, writes and refreshes the cells as it runs