Computer Memory Organization

The world has witnessed a gradual shift in data storage mechanisms over a period of years exterminating the use of data in printed form (hard copy) and moving towards a more versatile and useful form of data in storage devices. The amount of electronic data being exchanged world over has seen an exponential rise and has been streamlined and well-structured to increase its usability under various applications. The scale at which this data transfer process has reached is evident from the fact that it now supports the functioning of several autonomous organizations through the processes of data entry, its structuring and filtering, updation through large databases and integrated softwares.

So as we delve deep into the sea of data organization, the concept of memory and its management surfaces. Most of us have would have encountered the wide array of choices that are displayed in the marketplace when going on a hunt for a laptop/personal computer. Here we have our first brush with specifications like that of SDRAM, cache memory, hard disk drive along with the speed of transfer of FSB (Front Side Bus) and other accessories and add-ons. All the components specified above belong to the memory class but with varying sizes and speeds of operation. The most expensive, fastest memory is in close proximity to the processor. Memory contents are duplicated and expanded in banks, lower in the hierarchy. While the CPU registers and cache memory are the fastest in operation, they are expensive and small in size of the order of KBs for CPU registers and few MBs for cache memory. They store and access the most recent and frequently executed programs and enable easy and speedy reference to them. Random Access Memory (referenced precisely as Synchronous Dynamic RAM) stores the currently running program in memory and interacts with it, processing the various tasks defined under those set of instructions. RAM, as the name suggests allows random access to the storage location of data and hence is faster than the referencing mechanism used in Hard Disk Drives (HDD) which is Serial Advanced Technology Attachment (SATA). But on the contrary, the hard disks are of much larger capacity that is of the order of Gigabytes (80 GB, 120 GB, 160 GB, etc) and now even systems with Terabytes of storage have been rolled out by the hardware manufacturing firms.

Also, the hard disk consists of magnetic disks stacked one over the other to form a cylindrical structure. This structure is then divided into tracks and sectors which designate the location of data on the disk. Each disk contains one read and write head (R/W) for performing the associated function. In addition, there are some concepts that define the operational speed of the memory elements and help us analyze the responsiveness and application of the various available memories. ‘Hit rate’ is defined as the ratio of times a desired element is in that level of memory to the number of total accesses to memory. Further, ‘Hit time’ is the time to access the upper level of memory and deliver the item to the requesting device. Hence, according to the area of application of the device, its environment and the tasks associated with it, there can be a different combination of these memories for maximum efficiency. Judging from the pace at which the electronic data is penetrating and establishing control of our processes, it would not be uncanny to prognosticate that this branch of computers associated with memory management will have to re-discover itself time and again with path breaking technologies.

Ishant Arora

[Image Source]