Computing desk
< June 8	<< May \| June \| Jul >>	Current desk >

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

June 9

In an SSD, what's the physical route between the controller and the bit

In a HDD, there's a needle to read/write bits. The disk moves and the needle can reach any point of the disk.

In SSD memory type, how does the controller reach all bits? Are there cables? I imagine the memory block as a 3d matrix, with bit in the middle. Since there are no moving parts, some bits have to be behind others and not on the surface. Can someone link to a graphical explanation of this? Bumptump (talk) 21:56, 9 June 2023 (UTC)[reply]

Very simplistically, you can arrange the bits in a rectangular matrix, with a bit address (i,j) being selected by the line through the i-th row together with the line through the j-th column. When both lines are high, AND gates will select only bit (i,j). --Lambiam 09:56, 10 June 2023 (UTC)[reply]

For more detail, see Flash_memory#Principles_of_operation RudolfRed (talk) 19:13, 10 June 2023 (UTC)[reply]

That section is mainly about how the content of a selected cell is read and written in various technologies. It hardly addresses the actual question, how a cell is selected among all these cells. The text states, "the desired group is selected (in the same way that a single transistor is selected from a NOR array)", but for most readers this is not a helpful explanation. --Lambiam 22:36, 10 June 2023 (UTC)[reply]

Just to clear things up, an HDD is not like an LP record, there are no grooves on the platters, and no 'tracks': what you are calling a 'needle' is the double-sided read-write head which magnetically encodes and reads binary data on the platter. There is no contact whatever between the platter and the head - such contact is generally fatal to the microscopic tolerances involved. The HDD controller has a built-in map of every sector and it reads or writes data in sectors sequentially as requested via the operating system's low-level hardware interface routines. There's generally quite a lot of wasted space on an HDD, depending on how it's formatted and the type of data being written.

An SDD is much simpler: it's just a load of flash RAM on the end of a SATA bus, and it uses a standard Flash memory controller to read and write memory locations. This web page gives a general overview of things. This page has some low-down explanations. This page goes into more detail about individual controllers if you are interested. MinorProphet (talk) 22:05, 13 June 2023 (UTC)[reply]