Random-access memory

Random-access memory ( RAM / r æ m / ) is a form of computer data storage That stores data and computer code Currently being white used. A random-access memory device allows data items to be read or written in almost the same amount of time irrespective of the physical location of data inside the memory. In contrast, with other direct-access data storage media such as hard disks , CD-RWs , DVD-RWs and the older magnetic tapes and drum memoryThe time required to read and write data on the medium and the speed of motion.

RAM contains multiplexing and demultiplexing circuitry, to connect the data lines to the address entry. Usually, these devices are accessed by the same address, and RAM devices often have multiple data lines and are said to be “8-bit” or “16-bit”, and so on. devices.

In today’s technology, random-access memory takes the form of integrated circuits . RAM is normally associated with volatile types of memory (such as DRAM modules ), where stored information is lost if not volatile RAM has also been developed. [1] Other types of nonvolatile memories may exist, but they may be used for other purposes. These include most types of ROM and a type of flash memory called NOR-Flash .

Integrated circuit chip chips came into the market in the early 1970s, with the first commercially available chip chip, the Intel 1103 , introduced in October 1970. [2]


Early computers used relays , mechanical counters [3] or delay lines for main memory functions. Ultrasonic delay lines could only reproduce data in the order it was written. Drum memorycould be expanded at a relatively low cost cost effective retrieval of memory items. Latches built out of vacuum tube triodes , and later, out of discrete transistors , were used for smaller and faster memories such as registers. Such registers have been relatively large and too costly to use for large amounts of data; only a few bits or a few hundred bits of such memory could be provided.

The first practical form of random-access memory was the Williams tube starting in 1947. It stored data as electrically charged spots on the face of a cathode ray tube . Since the electron beam of the CRT could be read The capacity of the Williams tube was a few hundred to a thousand bits, but it was much smaller, faster, and more power-efficient than using individual vacuum tube latches. Developed at the University of Manchester in England, the Williams tube provided the medium on which the first electronically stored-memory program was implemented in the Manchester Small-Scale Experimental Machine(SSEM) computer, which first successfully ran a program on 21 June 1948. [4] In fact, rather than the Williams tube memory being designed for the SSEM, the SSEM was a testbed to demonstrate the reliability of the memory. [5] [6]

Magnetic-core memory was invented in 1947 and developed until the mid-1970s. It became a widespread form of random-access memory, relying on an array of magnetized rings. By changing the sense of each ringing magnetization, data could be stored with one bit stored per ring. Since every ring has had a combination of addresses to read and write, it has been possible.

Magnetic core memory was the standard form of memory system until it was released by solid-state memory in integrated circuits, starting in the early 1970s. Dynamic random-access memory(DRAM) is a replacement for a 4-bit or 6-transistor latch circuit by a single transistor for each memory bit, which increases the cost of volatility. Data was stored in the tiny capacitance of each transistor, and had to be periodically refreshed every few milliseconds before the charge could leak away. The Toshiba Toscal BC-1411 electronic calculator , which was introduced in 1965, [7] [8] used a form of DRAM built from discrete components. [8] DRAM was then developed by Robert H. Dennard in 1968.

Prior to the development of integrated read-only memory (ROM) circuits, permanent (or read-only ) random-access memory was often constructed using diode arrays driven by address decoders , or specially wound core rope memory planes. quote needed ]

Types of random-access memory

The two widely used RAMs are static RAM (SRAM) and dynamic RAM (DRAM). In SRAM, a bit of data is stored using the state of six transistor memory cells . This form of RAM is more expensive to produce, but it is faster and requires less dynamic power than DRAM. In modern computers, SRAM is often used as cache memory for the CPU. DRAM stores a bit of data using a transistor and capacitor pair, which together includes a DRAM cell. The capacitor holds a high or low load (1 or 0, respectively), and the transistor acts as a switch to the capacitor’s control of the capacitor’s state of charge or change it. As it is, it is the predominant form of computer memory used in modern computers.

Both static and dynamic RAM are considered volatile , and their state is lost when they are removed from the system. By contrast, read-only memory (ROM) data by permanently enabling or disabling selected transistors, such that the memory can not be altered. Writeable variants of ROM (Such as EEPROM and flash memory ) share properties of Both ROM and RAM, Enabling data to persist without power and to be updated without Requiring special equipment. These include USB flash drives, memory cards for portable devices, and solid-state drives . ECC memory(which can be either SRAM or DRAM) includes special circuitry to detect and / or correct random errors in the stored data, using parity bits or error correction codes .

In general, the term RAM refers to solid-state memory devices (or DRAM or SRAM), and more specifically the main memory in most computers. In optical storage, the term DVD-RAM is somewhat of a misnomer since, unlike CD-RW or DVD-RW it does not need to be erased before reuse. Nevertheless, a DVD-RAM behaves much like a hard drive drive if somewhat slower.

Memory cell

Main article: Memory cell (computing)

The memory cell is the fundamental building block of computer memory . The memory cell is an electronic circuit that stores one bit of binary information and it must be set to store a logic 1 (high voltage level) and reset to store a logic 0 (low voltage level). Its value is maintained / stored until it is changed by the set / reset process. The value in the memory can be accessed by reading it.

In SRAM, the memory cell is a type of flip-flop circuit, usually implemented using FETs . This means that SRAM requires very low power when not being accessed, but it is expensive and has low storage density.

A second type, DRAM, is based around a capacitor. Charging and discharging this capacitor can store a “1” or a “0” in the cell. However, the charge in this capacitor slowly leaks away, and must be refreshed periodically. Because of this refresh process, DRAM uses more power, but it can achieve greater storage densities and lower unit costs compared to SRAM.

Cell DRAM (1 Transistor and one capacitor)
SRAM Cell (6 Transistors)


To be useful, memory cells must be readable and writeable. Within the RAM device, multiplexing and demultiplexing circuitry is used to select memory cells. Typically, a device has a set of address lines A0 … An, and for each combination of bits that may be applied to these lines, a set of memory cells are activated. Due to this addressing, RAM devices almost always have a memory capacity that is a power of two.

Usually several memory cells share the same address. For example, a 4 bit ‘wide’ RAM chip has 4 memory cells for each address. Often the width of the memory of the microprocessor is different, for a 32 bit microprocessor, eight 4 bit RAM chips would be needed.

Often more addresses are needed than can be provided by a device. In that case, external multiplexors to the device are used.

Memory hierarchy

Main article: Memory hierarchy

One can read and over-write data in RAM. Many computer systems have a memory hierarchy consisting of processor registers , on-die SRAM caches, external caches , DRAM , paging systems and virtual memory or swap space on a hard drive. This entire pool of memory can be referred to as “RAM” by many developers, even though the various subsystems can be very different access times , violating the original concept behind the random access term in RAM. Even within a hierarchy level such as DRAM, the specific row, column, bank, rank , channel, or interleaveorganization of the components makes the access time variable, but not to the extent that access time to rotating storageis variable. The overall goal of using a memory is to obtain the highest possible average performance access while minimizing the total cost of the entire memory system (generally, the memory hierarchy follows the access time with the fast CPU registers at the top and the slow hard drive at the bottom).

In many modern personal computers, the RAM comes in an easily upgraded form of modules called memory modules or DRAM modules about the size of a few sticks of chewing gum. These can quickly be replaced by storage capacity. As suggested above, smaller amounts of RAM (mostly SRAM) are also integrated into the CPU and other ICs on the motherboard , as well as hard drives, CD-ROMs , and several other parts of the computer system.

Other uses of RAM

RAM is used in many other ways.

Virtual memory

Main article: Virtual memory

Most modern operating systems employ a method of extending RAM capacity, known as “virtual memory”. A portion of the computer’s hard drive is set aside for a paging file or a scratch partition , and the combination of physical RAM and the paging file system is total memory. (For example, if a computer has 2 GB of RAM and a 1 GB page, the operating system has 3 GB total memory available to it.) When the system runs low on physical memory, it can ” swap ” portions of RAM to the paging file to make room for new data, as well as read previously swapped information back to RAM. Excessive use of this mechanism results in thrashing and generally speaking, they are better than RAM.

RAM disk

Main article: RAM drive

Software can “partition” a portion of a computer’s RAM, allowing it to be much faster than RAM disk . A RAM disk loses the stored data when the computer is shut down, unless memory is provided to a standby battery source.

Shadow RAM

Sometimes, the contents of a relatively slow ROM are copied to read / write memory to allow for shorter access times. The ROM chip is then disabled while the initialized memory is switched over to the same block of addresses (often write-protected). This process, sometimes called shadowing , is fairly common in both computers and embedded systems .

As a common example, the BIOS has an option called “use shadow BIOS” or similar. When enabled, functions that rely on data from the BIOS ROMs instead of DRAM rentals (most can also be used to set the video card ROM or other ROM sections). Depending on the system, this may not result in increased performance, and may cause incompatibilities. For example, some hardware may be inaccessible to the operating system if RAM is used. Some systems can be hypothetical because the BIOS is not used after booting in favor of direct hardware access. Free memory is reduced by the size of the shadowed ROMs. [9]

Recent developments

Several new types of nonvolatile RAM , which preserve data while powered down, are under development. The technologies used include carbon nanotubes and approaches Utilizing tunnel magnetoresistance . Amongst the 1st generation MRAM , a 128 KiB ( 128 × 2 10 bytes) chip was manufactured with 0.18 μm technology in the summer of 2003. citation needed ] In June 2004, Infineon Technologies unveiled at 16 MiB (16 × 20 bytes) prototype based on 0.18 μm technology. There are two 2nd generation techniques currently in development:thermal-assisted switching (TAS) [10] which is being developed by Crocus Technology , and spin-transfer torque (STT) on which Crocus , Hynix , IBM , and several other companies are working. [11] Nantero built a functioning carbon nanotube 10 prototype prototype GiB (10 × 2 30 bytes) array in 2004. Whether some of these technologies can eventually take over significant market share DRAM, SRAM, gold flash-memory technology, however, remains to be seen.

Since 2006, ” solid-state drives ” (based on flash memory) with 256 gigabytes of capacity and performance have become available. This development has begun to blur the definition between traditional random-access memory and “disks”, dramatically reducing the difference in performance.

Some kinds of random-access memory, such as “EcoRAM”, are specifically designed for server farms , where low power consumption is more important than speed. [12]

Memory wall

The “memory wall” is the growing disparity of speed between CPU and memory outside the CPU chip. An important reason for this disparity is the limited communication bandwidth beyond chip boundaries, which is also referred to as the bandwidth wall . From 1986 to 2000, CPU speed was improved at an annual rate of 55% while memory speed only improved at 10%. Given these trends, it was expected that memory latency would become an overwhelming bottleneck in computer performance. [13]

CPU speed improvements slowed down significantly in the future. Intel summarized these causes in a 2005 document. [14]

“First of all, as chip geometries shrink and clock frequencies rise, the transistor leakage current increases, leading to excess power consumption and heat … Secondly, the advantages of higher clock speeds are in part negated by memory latency, since memory access times have not been able to keep pace with increasing clock frequencies. Third, for certain applications, traditional serial architectures are becoming more efficient and faster than Von Neumann bottleneck . In addition, in the field of inductance within solid state devices, resistance-capacitance (RC) delays in signal transmission are growing in the form of shrinkage.

The RC delays in signal transmission were also noted in the IPC: The End of the Road for Conventional Microarchitectures , which projected a maximum of 12.5% ​​average annual performance between 2000 and 2014.

A different concept is the processor-memory performance gap, which can be addressed by 3D integrated circuits that reduce the distance between the logic and memory aspects that are further apart in a 2D chip. [15] Memory subsystem design requires a focus on the gap, which is widening over time. [16] The main method of bridging the gap is the use of caches ; small amounts of high-speed memory, where they are called upon frequently. Multiple levels of caching have been developed with the widening gap, and the performance of high-speed modern computers is linked to evolving caching techniques.[17] These can prevent the loss of processor performance, as it takes the time to complete the computation. [18] There is 53% difference between the growth in speed of processor speeds and the lagging speed of main memory access. [19]

In contrast, RAM can be as fast as 5766 MB / s vs 477 MB / s for an SSD . [20]

See also

  • CAS latency (CL)
  • Hybrid Memory Cube
  • Multi-channel memory architecture
  • Registered / buffered memory
  • Parity RAM
  • Memory Interconnect / RAM nozzles
  • Memory geometry
  • Chip creep


  1. Jump up^ Gallagher, Sean. “Memory that never forgets: non-volatile DIMMs hit the market” . Ars Technica. Archived from the original on 2017-07-08.
  2. Jump up^ Bellis, Mary. “The Invention of the Intel 1103” .
  3. Jump up^ “IBM Archives – FAQ’s for Products and Services” . ibm.com . Archived from the original on 2012-10-23.
  4. Jump up^ Napper, Brian, Computer 50: The University of Manchester Celebrates the Birth of the Modern Computer , archived from the original on May 4 ,2012 , retrieved 26 May 2012
  5. Jump up^ Williams, FC; Kilburn, T. (Sep 1948), “Electronic Digital Computers”, Nature, 162 (4117): 487, doi : 10.1038 / 162487a0 . Reprinted inThe Origins of Digital Computers
  6. Jump up^ Williams, FC; Kilburn, T .; Tootill, GC (Feb 1951), “Universal High-Speed ​​Digital Computers: A Small-Scale Experimental Machine” , Proc. IEE , 98(61): 13-28, doi : 10.1049 / pi-2.1951.0004 , archived from the original on 2013-11-17.
  7. Jump up^ Toscal BC-1411 calculator Archived2017-07-29 at theWayback Machine.,Science Museum, London
  8. ^ Jump up to:b Toshiba “Toscal” BC-1411 Desktop Archived Calculator 2007-05-20 at the Wayback Machine .
  9. Jump up^ “Shadow Ram” . Archived from the original on 2006-10-29 . Retrieved 2007-07-24 .
  10. Jump up^ The Emergence of Practical MRAM “Archived copy” (PDF) . Archived from the original (PDF) on 2011-04-27 . Retrieved 2009-07-20 .
  11. Jump up^ “Tower invests in Crocus, tips MRAM foundry deal” . EETimes . Archived from the original on 2012-01-19.
  12. Jump up^ “EcoRAM held up as less power-hungry option than DRAM for server farms” Archived2008-06-30 at theWayback Machine. by Heather Clancy 2008
  13. Jump up^ The term was coined in “Archived copy” (PDF) . Archived (PDF) from the original on 2012-04-06 . Retrieved 2011-12-14 . .
  14. Jump up^ “Platform 2015: Intel® Processor and Evolution Platform for the Next Decade” (PDF) . March 2, 2005. Archived (PDF) from the original on April 27, 2011.
  15. Jump up^ Rainer Waser (2012). Nanoelectronics and Information Technology . John Wiley & Sons. p. 790. Archived from the original on August 1, 2016 . Retrieved March 31, 2014 .
  16. Jump up^ Chris Jesshope and Colin Egan (2006). Advances in Computer Systems Architecture: 11th Asia-Pacific Conference, ACSAC 2006, Shanghai, China, September 6-8, 2006, Proceedings . Springer. p. 109. Archived from the original on August 1, 2016 . Retrieved March 31, 2014 .
  17. Jump up^ Ahmed Amine Jerraya and Wayne Wolf (2005). Multiprocessor Systems-on-chips . Morgan Kaufmann. pp. 90-91. Archived from the original on August 1, 2016 . Retrieved March 31, 2014 .
  18. Jump up^ Impact of Advances in Computing and Communications Technologies on Chemical Science and Technology . National Academy Press. 1999. p. 110. Archived from the original on August 1, 2016 . Retrieved March 31,2014 .
  19. Jump up^ Celso C. Ribeiro and Simone L. Martins (2004). Experimental and Efficient Algorithms: Third International Workshop, WEA 2004, Angra Dos Reis, Brazil, May 25-28, 2004, Proceedings, Volume 3 . Springer. p. 529.Archived from the original on August 1, 2016 . Retrieved March 31, 2014 .
  20. Jump up^ Pinola, Melanie. “Add a RAM Disk to Your Computer for Faster-than-SSD Performance” . Lifehacker . Archived from the original on 10 September 2017 . Retrieved 10 September 2017 .

Leave a Reply

Your email address will not be published. Required fields are marked *

Copyright computerforum.eu 2019
Shale theme by Siteturner