MEMORY SYSTEMS

Memory systems store program information and other data in a specialized, high-speed component that facilitates efficient access. To accomplish this task, various technologies and organizational methods are employed. The typical computer system is equipped with a hierarchy of memory subsystems; some (internal to the system) can be directly accessed by the processor and some can be accessed by the processor via I /O devices (external). Internal memory consists of main memory, cache, and registers (the CPU’s own local memory).
External memory consists of secondary storage devices, such as magnetic disk and tape, and optical memory storage on CD-ROM. One of the most fundamental components of a PC is its memory, and there are several
components that go into the memory of a PC. Terms such as random access memory (RAM), dynamic RAM (DRAM), read-only memory (ROM), virtual memory, flash memory, erasable programmable ROM (EPROM), and many more all refer to memory. The word component is used loosely due to the fact that while most memory is a physical object (created on a microchip), it can be created virtually through the use of the hard-disk drive or other unused media. However, physical memory is always faster than
virtual memory, by definition. This is the reason why computers with a lot of physical
memory are faster than computers with a lot of virtual memory.
Memory can be found in a wide variety of common everyday items where information
from the user (or manufacturer) needs to be stored. These can range from computers, to
cellular phones, to car radios, to VCRs. Of course, this information is only stored until the
power source is removed from the memory.
There are two main categories of memory—volatile and nonvolatile—and within these
two categories exists a large number of subcategories. Volatile memory is characterized by
the fact that the information stored in the memory is only retained as long as there is a
constant electricity flow to the memory. Once the energy flow is shut off, the stored information
is lost. Computer RAM is a primary example of volatile memory as it is the primary
memory used in a computer.
Nonvolatile memory, the counterpart to volatile memory, retains its information even
when the flow of electricity is disrupted. This is the case for both physical and virtual
memory. Within nonvolatile memory, there are two further classifications: permanent, where
it can only be written once (read only), and erasable, where the memory can be reused. A
good example of permanent nonvolatile memory is the ROM chips where the startup program
and the BIOS are stored. Examples of erasable nonvolatile memory are flash memory, memory
sticks, and smart media.
Technically, the first example of nonvolatile memory came about between 1948 and
1949 when Jay Forrester and other individuals in various locations developed core memory.
The basic premise for core memory is that small iron doughnuts would have wires passed
through them that would be able to carry a magnetic charge. These charges, either positive
or negative, could represent either 0 or 1 states and would be transferred from the wires into
the doughnuts. What was significant about this was that these doughnuts retained their charge
even after the current to the wires was stopped. Thus the first nonvolatile memory was
created.
The issue of combining temporary and permanent storage seemed to have found a
solution in 1956 when IBM released its first hard drive. However, this was not the case. In
1958 the microchip, or integrated circuit, was demonstrated for the first time, which ultimately
led to the foundation of many memory devices today. It was through the use of these
circuits that two different states, 0 and 1, could be created and presented a great possibility
for both permanent and temporary storage.
It was not until June 1968, however, that the first patent was granted to Robert Denard
and IBM for a one-transistor DRAM cell. During the 10 years previous to this, the only
applications of the microchip had been as a miniature engine inside of hearing aids, missiles,
and eventually calculators. Denard’s (and IBM’s) DRAM was a taste of what would become
much more complex memory chips.
From 1969 to 1971, the first generation of memory chips began to appear in both
calculators and computers. First, Intel released its 1-kB RAM chip, a powerful one at the
time, followed by its first commercially available DRAM chip, the Intel 1103, the following
year. That same year (1970), Canon incorporated a Texas Instruments RAM chip into one
of its calculators, soon to follow with a ROM chip in its Canon L100A and LE-10 calculators.
From these humble beginnings, memory chips have grown in speed and power, though the
fundamentals of the chips are still the same.
RAM is random-access memory because it appropriately utilizes byte storage in more
of a random manner rather than a temporary one. In particular, any piece of storage can be
used regardless of order and will not disturb any of the other bytes already stored there.
The process of writing and erasing in RAM is an automatic process that occurs many
times during the device’s operation. As new information is stored and used, other information
will be erased and removed to free up memory for other applications. However, RAM cannot
be manually erased at will, like a hard-disk drive or cassette tape. Nonetheless, the effects
of erasing can be seen when the user instructs the device to erase a particular programming,
such as the programmed recording time on a VCR.
There are two basic categories of RAM: DRAM and SRAM (static RAM). Since the
1970s, DRAM has helped run computers and continues to do so today. Compared to the
newer SRAM, however, it can be up to 10 times slower. The reason for this is that DRAM
requires that it be refreshed (recharged with electricity) several thousands of times per second
whereas SRAM only needs to be refreshed whenever there is a write procedure. The time
difference between these two different types of RAM is 5 ns (nanoseconds) versus 30–50
ns for the DRAM. A significant factor in the use of SRAM stems from the fact that in
addition to being faster than DRAM, it is significantly more expensive. Because of this,
DRAM is still the most popular kind of memory used in computers today. This means that
SRAM is typically reserved for caches, which are naturally smaller (approximately 2 MB)
and other electronic devices that do not require a lot of memory.
The first DRAM memory only worked at speeds of 300 ns, and as such, early computer
processors were faster than their memory, wasting the potential utilization of the CPU. However,
as processors have become faster, the increases in memory speed have not been able
to keep up with the increases in CPU sppeds. As such, there has been an abundance of
demand for faster DRAM.
To get around this issue, manufacturers have devised several solutions to help alleviate
the problem. Some solutions are caching, as mentioned before, synchronizing with the computer’s
clock, bursting (receiving and processing multiple data addresses at once), and data
rate doubling (data are sent twice during a clock cycle rather than once). Some variations
of DRAM that use some or many of these techniques are EDO RAM (extended data output
RAM), BEDO RAM (burst extended data output RAM), SDRAM (synchronous dynamic
RAM), and DDR DRAM (double-data-rate SDRAM).
In addition to ROM, another type of permanent nonvolatile memory that needs to be
mentioned is PROM (programmable ROM). PROM is similar to a CD in that the information
is burned into the memory after it has been manufactured (as opposed to ROM where the
information is incorporated while the memory is being manufactured).
The first PROM was invented by Texas Instruments in the late 1970s and was quickly
hailed as a better version of memory than ROM because manufacturers would wait until the
last moment to input their latest BIOS. In doing so, the manufacturers helped to maintain
the shelf life of their product. However, even with this advantage, older BIOSs would be
rendered obsolete when the newer ones were released, driving manufacturers to search for
a reprogrammable BIOS chip.
The answer to the manufacturers’ requests came with the advent of the EPROM (erasable
PROM)—a special kind of PROM that would erase when subjected to ultraviolet light.
Utilizing special equipment, the EPROM chip could then be reprogrammed. However, of
even more significance than the EPROM was the creation of EEPROM (electrically erasable
PROM)—another special case of PROM that could be erased with an electrical charge.
From the development of EEPROM came many of the flash memory, Smart Media, and
Compact Flash based memory devices that all involve nonvolatile, erasable, and reprogrammable
memory chips. A large majority of this technology can be found either installed or
as an external component of digital cameras and audio players.

READ MORE.......

Emory W. Zimmers, Jr. and Technical Staff
Enterprise Systems Center
Lehigh University
Bethlehem, Pennsylvania




for STEP BY STEP GUIDE autocad simple tutorial please visit.........
www.autocadsimpletutorial.blogspot.com


---or---

www.autocad-simple-tutorial.com

Berbagi :