Memory Organization

Memory Basics Memory is the workspace for the computers processor. It is a temporary storage area where the programs and data being operated on by the processor must reside. Memory storage is considered temporary because the data and programs will remain there only as long as the computer has electrical power or is not reset. Before being shut down or reset, any data that has been changed should be saved to a more permanent storage device of some type (usually a hard disk) so it can be reloaded into memory again in the future. We often call memory RAM, for Random Access Memory.Main memory is called RAM because you can randomly (and quickly) access any location in memory. When we talk about a computers memory, we usually mean the RAM in the system, meaning primarily the memory chips or modules that make up the primary active program and data storage used by the processor. This is often confused with the term storage, which should be used when referring to things such as disk and tape drives (although some people do consider them a form of memory). 2Types of Memory To better understand physical memory in a system, it is necessary to see where and how it fits into the system.Three main types of physical memory used in modern PCs are †¢ ROM: Read Only Memory †¢ DRAM: Dynamic Random Access Memory †¢ SRAM: Static RAM 2. 1ROM Read Only Memory, or ROM, is a type of memory that can permanently or semipermanently hold data. It is called read-only because it is either impossible or difficult to write to. ROM is also often referred to as non-volatile memory because any data stored in ROM will remain, even if the power is turned off. As such, ROM is an ideal place to put the PCs startup instructionsthat is, the software that boots the system.Note that ROM and RAM are not opposites, as some people seem to believe. In fact, ROM is technically a subset of the systems RAM. In other words, a portion of the systems Random Access Memory address space is mapped into one or more ROM chips. This is necessary to contain the software that enables the PC to boot up; otherwise, the proces- sor would have no program in memory to execute when it was powered on. For example, when a PC is turned on, the processor automatically jumps to address FFFF0h, expecting to find instructions to tell the processor what to do.This location is exactly 16 bytes from the end of the first megabyte of RAM space, and the end of the ROM. If this location was mapped into regular memory 1 chips, any data stored there would have disappeared when the power was turned off previously, and the processor would subsequently find no instructions to run the next time power was turned on. By placing a ROM chip at this address, a system startup program can be permanently loaded into the ROM and will be available every time the system is turned on. The motherboard ROM normally contains four main programs, including the following in most systems: †¢ POST: Power-On Self Test.A series of test routines that ensure the system components are operating properly. †¢ CMOS Setup: A menu-driven application that allows the user to set sys- tem configuration parameters, options, security settings, and preferences. †¢ Bootstrap Loader: The routine that first scans the floppy drive and then the hard disk, looking for an operating system to load. †¢ BIOS: Basic Input/Output System. A series of device driver programs designed to present a standard interface to the basic system hardware, especially hardware that must be active during the boot process. Four different types of ROM chips are ROM. Read Only Memory †¢ PROM. Programmable ROM †¢ EPROM. Erasable PROM †¢ EEPROM. Electrically Erasable PROM, also called a flash ROM No matter which type of ROM you use, the data stored in a ROM chip is non- volatile and will remain indefinitely unless intentionally erased or overwritten. 2. 1. 1PROM PROMs are a type of ROM that is blank when new and must be pr ogrammed with whatever data you want 2. 1. 2EPROM One variation of the PROM that has been very popular is the EPROM. An EPROM is a PROM that is erasable. EPROM is erased by exposure to intense UV light. 2. 1. 3 EEPROM/Flash ROMA newer type of ROM is the EEPROM, which stands for Electrically Erasable PROM. These chips are also called flash ROMs, and are characterized by their capability to be erased and reprogrammed directly in the circuit board in which they are installed, with no special equipment required. 2 2. 2DRAM Dynamic RAM is the type of memory chip used for most of the main memory in a modern PC. The main advantages of DRAM is that it is very dense, meaning you can pack a lot of bits into a very small chip, and it is very inexpensive, which makes it affordable for large amounts of memory.The memory cells in a DRAM chip are tiny capacitors that retain a charge to indicate a bit. The problem with DRAM is that it is dynamic, and because of the design must be constantly refresh ed or the electrical charges in the individual memory capacitors will drain and the data will be lost. Refresh occurs when the system memory controller takes a tiny break and accesses all the rows of data in the memory chips. DRAMs use only one transistor and capacitor pair per bit, which makes them very dense, offering a lot of memory capacity per chip than other types of memory. 2. 3 Cache MemorySRAMThere is another distinctly different type of memory that is significantly faster than most types of DRAM. SRAM stands for Static RAM, which is so named because it does not need the periodic refresh rates like DRAM (Dynamic RAM). Due to the design of SRAM, not only are refresh rates unnecessary, but SRAM is much faster than DRAM and is fully able to keep pace with modern processors. SRAM memory is available in access times of 2ns or less, which means it can keep pace with processors running 500MHz or faster! This is due to the SRAM design, which calls for a cluster of six transistors f or each bit of storage.The use of transistors but no capacitors means that refresh rates are not necessary because there are no capacitors to lose their charges over time. As long as there is power, SRAM will remember what is stored. Compared to DRAM, SRAM is much faster, but also much lower in density and much more expensive. The lower density means that SRAM chips are physically larger and store many less bits overall. The high number of transistors and the clustered design means that SRAM chips are both physically larger and much more expensive to produce than DRAM chips.Even though SRAM is too expensive for PC use as main memory, PC designers have found a way to use SRAM to dramatically improve PC performance. Rather than spend the money for all RAM to be SRAM memory, which can run fast enough to match the CPU, it is much more cost-effective to design in a small amount of high-speed SRAM memory, called cache memory. The cache runs at speeds close to or even equal to the processo r, and is the memory from which the processor normally directly reads from and writes to. During read operations, the data in the high-speed cache memory is resupplied from the lower-speed main memory or DRAM in advance. 3Memory Packaging Memory is made from tiny semiconductor chips and must be packaged into something less fragile and tiny in order to be integrated with the rest of the system Different types of memory paclages are †¢ Dual Inline Packages (DIPs) and Memory Modules †¢ Single Inline Memory Modules (SIMMs) †¢ Dual Inline Memory Modules (DIMMs) 3. 1DIPS Most memory chips are packaged into small plastic or ceramic packages called dual inline packages or DIPs. A DIP is a rectangular package with rows of pins running along its two longer edges These are the small black boxes you see on SIMMs, DIMMs or other larger packaging styles . 2SIMMs SIMMs are available in two flavors: 30 pin and 72 pin. 30-pin SIMMs are the older standard, and were popular on third and fourth generation motherboards. 72-pin SIMMs are used on fourth, fifth and sixth generation PCs. SIMMs are placed into special sockets on the motherboard created to hold them. The sockets are specifically designed to ensure that once inserted, the SIMM will be held in place tightly 3. 3DIMMs DIMMs are 168 pins in size, and provide memory 64 bits in width. They are a newer form factor and are becoming the de facto standard for new PCs; they are not used on older motherboards motherboards.SIMMs have contacts on either side of the circuit board but they are tied together. So a 30-pin SIMM has 30 contacts on each side of the circuit board, but each pair is connected. DIMMs however have different connections on each side of the circuit board 4Memory Banks Memory chips (DIPs, SIMMs, SIPPs, and DIMMs) are organized in banks on motherboards and memory cards. The banks usually correspond to the data bus capacity of the system’s microprocessor. The number of bits for each bank can be made up of single chips, SIMMs, or DIMMs. 4 5Memory ReliabilityA part of the nature of memory is that it will inevitably fail. These failures are usually classified as two basic types: hard fails and soft errors. The most well understood are hard fails, in which the chip is working and then, due to some flaw, physical damage, or other event, becomes damaged and experiences a permanent failure. Fixing this type of failure normally requires replacement of some part of the memory hardware, such as the chip, SIMM, or DIMM. Hard error rates are known as HERs. The other more insidious type of failure is the soft error.A soft error is a nonpermanent failure that may never reoccur, or occur at infrequent intervals. (Soft fails are effectively â€Å"fixed† by powering the system off and back on. ) Soft error rates are known as SERs. There are basically three levels and techniques for fault tolerance used in modern PCs: †¢ Non-parity †¢ Parity †¢ ECC (Error Correcting Co de) Non-parity systems have no fault tolerance at all. The reason they are even used is because they have the lowest inherent cost. No additional memory is necessary as is the case with parity or ECC techniques. 6ParityTne standard IBM set for the industry is that the memory chips in a bank of nine each handle one bit of data: eight bits per character plus one extra bit called the parity bit. As the eight individual bits in a byte are stored in memory, a parity genera- tor/checker, which is either part of the CPU or located in a special chip on the motherboard, evaluates the data bits by counting the number of 1s in the byte. If an even number of 1s is in the byte, the parity generator/checker creates a 1 and stores it as the ninth bit (parity bit) in the parity memory chip.That makes the total sum for all nine bits an odd number. If the original sum of the eight data bits is an odd number, the parity bit created is 0, keeping the 9-bit sum an odd number. The value of the parity bit is always chosen so that the sum of all nine bits (eight data bits plus one parity bit) is an odd number. The following examples may make it easier to understand: Data bit number: 0 1 2 3 4 5 6 7 Data bit value:10110011 Parity bit:0 5 In this example, because the total number of data bits with a value of 1 is an odd number (5), the parity bit must have a value of 0 to ensure an odd sum for all nine bits.The following is another example: Data bit number: 0 1 2 3 4 5 6 7 Data bit value :00110011 Parity bit: 1 In this example, because the total number of data bits with a value of 1 is an even number (4), the parity bit must have a value of 1 to create an odd sum for all nine bits. When the system reads memory back from storage, it checks the parity information. If a (9-bit) byte has an even number of bits with a parity bit value of 1, that byte must have an error. The system cannot tell which bit has changed, or if only a single bit has changed.If three bits changed, for example, the byte still flags a parity-check error; if two bits changed, however, the bad byte may pass unnoticed. The following examples show parity-check messages for three types of systems: For the IBM PC: PARITY CHECK x For the IBM XT: PARITY CHECK x yyyyy (z) For the IBM AT and late model XT: PARITY CHECK x yyyyy Where x is 1 or 2: 1=Erroroccurredonthemotherboard 2=Erroroccurredinanexpansionslot yyyyy represents a number from 00000 through FFFFF that indicates, in hexadecimal notation, the byte in which the error has occurred. Where (z) is (S) or (E ): S) = Parity error occurred in the system unit (E ) = Parity error occurred in the expansion chassis When a parity-check error is detected, the motherboard parity-checking cir- cuits generate a non-maskable interrupt (NMI), which halts processing and di- 6 verts the systems attention to the error. The NMI causes a routine in the ROM to be executed. The routine clears the screen and then displays a message in the upper-left corner of the screen . The message differs depending on the type of computer system. 7 ECC (Error Correcting Code) ECC goes a big step beyond simple parity error detection.Rather than just detecting an error, ECC allows a single bit error to be corrected, which means the system can continue on without interruption and without corrupting data. ECC as implemented in most PCs can only detect and not correct double-bit errors. Because studies have indicated that approximately 98 percent of memory errors are single- bit variety, the most commonly used type of ECC is one in which the attendant memory controller detects and corrects single-bit errors in an accessed data word (double-bit errors can be detected, but not corrected).This type of ECC is known as SEC-DED and requires an additional seven check bits over 32 bits in a 4-byte system and eight check bits in an 8-byte system. ECC in a 4-byte system obviously costs more than non-parity or parity, but in an 8-byte system, ECC and parity costs are equal. ECC entails the memory controller calculating the check bits on a memory- write operation, performing a compare between the read and calculated check- bits on a read operation and, if necessary, correcting bad bit(s).The additional ECC logic in the memory controller is not very significant in this age of inex- pensive, high-performance VLSI logic, but ECC actually affects memory perfor- mance on writes. This is because the operation must be timed to wait for the calculation of check bits and, when the system waits for corrected data, reads. On a partial-word write, the entire word must first be read, the affected byte(s) rewritten, and then new check bits calculated. This turns partial-word write operations into slower read-modify writes. Most memory errors are of a single-bit nature, which are correctable by ECC.Incorporating this fault-tolerant technique provides high system reliability and attendant availability. An ECC-based system is a good choice for servers, workstations, or mis sion-critical applications in which the cost of a potential memory error outweighs the additional memory and system cost to correct it, along with ensuring that it does not detract from system reliability. 8 The System Logical Memory Layout The original PC had a total of 1M of addressable memory, and the top 384K of that was reserved for use by the system.Placing this reserved space at the top (between 640K and 1024K instead of at the bottom, between 0K and 640K) led to what today is often called the conventional memory barrier. The constant pressures on system and peripheral manufacturers to maintain compatibility by never breaking from the original memory scheme of the first PC has resulted in 7 a system memory structure that is (to put it kindly) a mess. Logical memory sections are given below †¢ Conventional (Base) memory †¢ Upper Memory Area (UMA) †¢ High Memory Area (HMA) †¢ Extended memory (XMS) †¢ Expanded memory (obsolete) Video RAM memory (part of UMA) †¢ Adapter ROM and Special Purpose RAM (part of UMA) †¢ Motherboard ROM BIOS (part of UMA) 8. 1 Conventional (Base) Memory The original PC/XT-type system was designed to use 1M of memory workspace, sometimes called RAM (random access memory). This 1M of RAM is divided into several sections, some of which have special uses. DOS can read and write to the entire megabyte, but can manage the loading of programs only in the portion of RAM space called conventional memory, which was 512K at the time the first PC was introduced.The other 512K was reserved for use by the system, including the motherboard and adapter boards plugged into the system slots. After introducing the system, IBM decided that only 384K was needed for these reserved uses, and the company began marketing PCs with 640K of user memory. Thus, 640K became the standard for memory that can be used by DOS for running programs, and is often termed the 640K memory barrier. The remaining memory after 640K was rese rved for use by the graphics boards, other adapters, and the motherboard ROM BIOS.This barrier largely affects 16-bit software such as DOS and Windows 3. 1, and is much less of a factor with 32-bit software and operating systems such as Windows 95/98, NT, and so on. 8. 2 Upper Memory Area (UMA) The term Upper Memory Area (UMA) describes the reserved 384K at the top of the first megabyte of system memory on a PC/XT and the first megabyte on an AT-type system. This memory has the addresses from A0000 through FFFFF. The way the 384K of upper memory is used breaks down as follows: †¢ The first 128K after conventional memory is called video RAM.It is re- served for use by video adapters. When text and graphics are displayed onscreen, the electronic impulses that contain their images reside in this space. Video RAM is allotted the address range from A0000-BFFFF. 8 †¢ The next 128K is reserved for the adapter BIOS that resides in read-only memory chips on some adapter boards plug ged into the bus slots. Most VGA-compatible video adapters use the first 32K of this area for their onboard BIOS. The rest can be used by any other adapters installed. Many network adapters also use this area for special-purpose RAM called Shared Memory.Adapter ROM and special-purpose RAM is allotted the address range from C0000-DFFFF. †¢ The last 128K of memory is reserved for motherboard BIOS (the basic input/output system, which is stored in read-only RAM chips or ROM). The POST (Power-On Self Test) and bootstrap loader, which handles your system at bootup until the operating system takes over, also reside in this space. Most systems only use the last 64K (or less) of this space, leaving the first 64K or more free for remapping with memory managers. Some systems also include the CMOS Setup program in this area.The motherboard BIOS is allotted the address range from E0000-FFFFF. 8. 3Extended Memory The memory map on a system based on the 286 or higher processor can extend bey ond the 1M boundary that exists when the processor is in real mode. On a 286 or 386SX system, the extended memory limit is 16M; on a 386DX, 486, Pentium, Pentium MMX, or Pentium Pro system, the extended memory limit is 4G (4,096M). Systems based on the Pentium II processor have a limit of 64G (65,536M). For a system to address memory beyond the first megabyte, the processor must be in protected modethe native mode of 286 and higher processors.On a 286, only programs designed to run in protected mode can take advantage of extended memory. 386 and higher processors offer another mode, called virtual real mode, which enables extended memory to be, in effect, chopped into 1M pieces (each its own real-mode session). Virtual real mode also allows for several of these sessions to be running simultaneously in protected areas of memory. The extended memory specification (XMS) was developed in 1987 by Mi- crosoft, Intel, AST Corp. , and Lotus Development to specify how programs would use exte nded memory.The XMS specification functions on systems based on the 286 or higher and allows real-mode programs (those designed to run in DOS) to use extended memory and another block of memory usually out of the reach of DOS. Before XMS, there was no way to ensure cooperation between programs that switched the processor into protected mode and used extended memory. There was also no way for one program to know what another had been doing with the extended memory because none of them could see that memory while in real mode. HIMEM.SYS becomes an arbitrator of sorts that first grabs all the extended memory for itself and then doles it out to programs that know the XMS protocols. In this manner, several programs that use XMS memory can operate together under DOS on the same system, switching the pro- cessor into 9 and out of protected mode to access the memory. Extended memory can be made to conform to the XMS specification by installing a de- vice driver in the CONFIG. SYS file. The most common XMS driver is HIMEM. SYS, which is included with Windows 3. x and later versions of DOS, starting with 4. and up. 8. 4 High Memory Area (HMA) The High Memory Area (HMA) is an area of memory 16 bytes short of 64K in size, starting at the beginning of the first megabyte of extended memory. It can be used to load device drivers and memory-resident programs to free up conventional memory for use by real-mode programs. Only one device driver or memory-resident program can be loaded into HMA at one time, no matter what its size. Originally, this could be any program, but Microsoft decided that DOS could get there first, and built capability into DOS 5 and newer versions.The HMA area is extremely important to those who use DOS 5 or higher because these DOS versions can move their own kernel (about 45K of program instructions) into this area. This is accomplished simply by first loading an XMS driver (such as HIMEM. SYS) and adding the line DOS=HIGH to your CONFIG. SYS file. Tak ing advantage of this DOS capability frees another 45K or so of conventional memory for use by real-mode programs by essentially mov- ing 45K of program code into the first segment of extended memory.Although this memory was supposed to be accessible in protected mode only, it turns out that a defect in the design of the original 286 (which, fortunately, has been propagated forward to the more recent processors as a feature) accidentally al- lows access to most of the first segment of extended memory while still in real mode. The use of the HMA is controlled by the HIMEM. SYS or equivalent driver. The origins of this memory usage are interesting because they are based on a bug in the original 286 processor carried forward through even the Pentium II. 8. 5 Expanded MemorySome older programs can use a type of memory called Expanded Memory Spec- ification or EMS memory. Unlike conventional (the first megabyte) or extended (the second through 16th or 4,096th megabytes) memory, expanded memory is not directly addressable by the processor. Instead, it can only be accessed through a 64K window and small 16K pages established in the UMA. Expanded memory is a segment or bank-switching scheme in which a custom memory adapter has a large number of 64K segments onboard, com- bined with special switching and mapping hardware. The system uses a free segment in the UMA as the home address for the EMS board.After this 64K is filled with data, the board rotates the filled segment out and a new, empty segment appears to take its place. In this fashion, you have a board that can keep on rotating in new segments to be filled with data. Because only one segment can be seen or oper- ated on at one time, EMS is very inefficient for program code and is normally 10 only used for data. 9Video Memory The video memory is such an important component of the video card, and indirectly the entire PC, that several new memory technologies have been created specifically for it.The goal: to impr ove the speed with which information can be pumped into and out of the video memory, to keep system performance high as the video system tries to do more and more. Various memory technologies now being used on video cards are explained below. 9. 1 Standard (Fast Page Mode) DRAM The oldest technology used in video card memory, fast page mode (FPM) memory is now considered â€Å"standard† DRAM as it has the fewest performanceenhancing capabilities of the different types of memory on the market. FPM DRAM is a technology used primarily for main system memories (even there, it is now considered a poor performer) and is not really ell-suited for highperformance video applications. 11 FPM is the least expensive type of memory available for video, and is used today mostly on low-end or generic cards (as well as older cards of course). For many applications they can be quite satisfactory; however, they reach their limits quickly when trying to use high resolution modes, especially in true color. The limitations of standard DRAM are due to two primary effects: it is single ported (which means it can only do one access at a time) and it runs at a relatively low speed and access width. 9. 2 Extended Data Out (EDO) DRAMEDO DRAM is the same as standard FPM DRAM except for a slight modification in the access cycle that gives it a small performance boost. With EDO DRAM, one read to memory can begin before the last one has completely finished; this yields a raw speed improvement of between 5 and 20 percent, depending on whom you ask. Originally used only for main system memory, EDO DRAM is becoming more popular on video cards because it provides slightly improved performance over standard DRAM at the same cost. (At one time EDO was more expensive than FPM but due to supply and demand effects now, EDO is actually the same cost or lower).EDO is still, however, a low-cost and low-performance solution compared to other types of video memory, and is not used on high-end card s. 9. 3 Video RAM (VRAM) The traditional, standard DRAM used for video cards typically does not have enough bandwidth to handle the demands of running a card at high resolution and color depths, with acceptable refresh rates. The main reason why is the two competing access factors for the video memory: the processor writing new information to the memory, and the RAMDAC reading it many times per second in order to send video signals to the monitor.To address this fundamental limitation, a new type of memory was created called video RAM or VRAM. As the name implies, this memory is specifically tailored for use in video systems. The fundamental difference between VRAM and standard DRAM is that VRAM is dual-ported. This means that it has two access paths, and can be written to and read from simultaneously. The advantages of this are of course enormous given what the video card does: many times per second a new screen image is calculated and written to the memory, and many times per seco nd this memory is read and sent to the monitor.Dual- porting allows these operations to occur without bumping into each other. VRAM provides substantially more bandwidth than either standard DRAM or EDO DRAM; double in many cases. It is more suited for use in systems requiring high resolution and color depth displays. The only reason that it hasn’t replaced standard DRAM entirely is of course: cost. VRAM is more complex and requires more silicon per bit than standard DRAM, which makes it cost more. 12 9. 4Window RAM (WRAM)Window RAM or WRAM is a modification of regular VRAM that both improves performance and reduces cost on a bit-for-bit basis. Designed specifically for use in graphics cards, WRAM is also dual-ported but has about 25% more bandwidth than VRAM, and also incorporates additional features to allow for higher performance memory transfers for commonly used graphical operations such as text drawing and block fills. Furthermore, WRAM is less expensive than VRAM to ma nufacture (although still more expensive than DRAM). 9. 5 Synchronous Graphics RAM (SGRAM)A relatively newer RAM technology, Synchronous Graphics RAM or SGRAM tackles the poor performance of regular DRAM by increasing greatly the speed at which memory transfers take place. SGRAM also incorporates specific per- formance enhancing features designed to work with acceleration features built into video cards, to greatly improve overall video processing speed. SGRAM is still single-ported, unlike VRAM or WRAM, but offers performance that is much closer to VRAM than DRAM due to its advanced design. 10 Flash Memory DevicesFlash memory has been around for several years as a main or an auxiliary storage medium for notebook computers. However, the rise of devices such as digital cameras and MP3 players and the presence of USB ports on practically all recent systems have transformed this technology from a niche product into a mainstream must-have storage technology. Flash memory is a type of no nvolatile memory that is divided into blocks rather than bytes, as with normal RAM memory modules. Flash memory, which also is used in most recent computers for BIOS chips, is changed by a process known as Fowler-Nordheim tunneling.This process removes the charge from the floating gate associated with each memory cell. Flash memory then must be erased before it can be charged with new data. The speed, low reprogramming current requirements, and compact size of recent flash memory devices have made flash memory a perfect counterpart for portable devices such as notebook computers and digital cameras, which often refer to flash memory devices as so-called â€Å"digital film†. Unlike real film, digital film can be erased and reshot.Ultra-compact, USB-based keychain drives that use flash memory are replacing both traditional floppy drives and Zip/SuperDisk drives for transporting data between systems. Diiferent types of flash memory devices are expained below. 10. 1Compact Flash CompactFlash was developed by SanDisk Corporation in 1994 and uses ATA architecture to emulate a disk drive; a CompactFlash device attached to a com- 13 puter has a disk drive letter just like your other drives. Later types of flash memory also use ATA architecture, either implemented in the device itself or in its controller. 0. 2MultiMedia Card The MultiMediaCard (MMC) was codeveloped by SanDisk and Infineon Tech- nologies AG (formerly Siemens AG) in November 1997 for use with smart phones, MP3 players, digital cameras, and camcorders. The MMC uses a simple 7-pin serial interface to devices and contains low-voltage flash memory. 10. 3Secure Digital (SD) A SecureDigital (SD) storage device is about the same size as MMC , but it’s a more sophisticated product. SD, which was codeveloped by Toshiba, Matsushita Electric (Panasonic), and SanDisk in 1999, gets its name from two special features.The first is encrypted storage of data for additional security, meeting current and fut ure Secure Digital Music Initiative (SDMI) standards for portable devices. The second is a mechanical write-protection switch. 10. 4Pen Drive As an alternative to floppy and Zip/SuperDisk-class removable-media drives, USB-based flash memory devices are rapidly becoming the preferred way to move data between systems. The first successful drive of this type – Trek’s ThumbDrivewas – introduced in 2000 and has spawned many imitators, in- cluding many that incorporate a keychain or pocket clip to emphasize their portability.Unlike other types of flash memory, USB keychain drives don’t require a separate card reader; they can be plugged in to any USB port or hub. Al- though a driver is usually required for Windows 98 and Windows 98SE, most USB keychain drives can be read immediately by newer versions of Windows, particularly Windows XP. As with other types of flash memory, USB keychain drives are assigned a drive letter when connected to the computer. Most have capacities ranging from 128MB to 1GB, with some capacities as high as 2GB or more.However, typical read/write performance of USB 1. 1-compatible drives is about 1MBps. Hi-Speed USB keychain drives are much faster, providing read speeds ranging from 5MBps to 15MBps and write speeds ranging from 5MBps to 13MBps. 11Advanced Memory Technologies 11. 1RDRAM RDRAM is a proprietary technology made by Rambus Inc. for use exclusively in certain Intel compatible motherboards 14 RDRAM stands for Rambus Dynamic Random Access Memory. It can access data anywhere on the chip; It requires power to hold its data; and it transfers data twice per clock signal.However, it uses a smaller pathway, or â€Å"system bus,† to send information. The Rambus system bus is 16-bits wide. Rambus transfers data at 800 megahertz (MHz) and faster. Rambus is the more expensive type of memory since its proprietary, royalty costs that manufacturers must pay to produce it tend to result in higher retail prices. Add itionally, Rambus compatible motherboards require that all of their RAM slots be occupied. Traditionally, a computer may contain anywhere from one to four ram slots. If a slot is unoccupied, the system still operates.Rambus requires that either a Rambus memory module or a kind of place holder known as a continuity module be in place to complete the memory path to the bus. 11. 2 DDR SDRAM (DDR) Double data rate (DDR) SDRAM memory is a JEDEC-created standard that is an evolutionary upgrade of standard SDRAM in which data is transferred twice as quickly. Instead of doubling the actual clock rate, DDR memory achieves the doubling in performance by transferring twice per transfer cycle: once at the leading (falling) edge and once at the trailing (rising) edge of the cycle.This effectively doubles the transfer rate, even though the same overall clock and timing signals are used. Since its inception, manufacturers have release new and faster versions of DDR. These are based on the use of p refetch buffers that access not only the memory, or â€Å"dataword,† requested by the processor but also the datawords adjacent to it on the chip. Thus DDR2 â€Å"fetches† four datawords per memory access, double the amount of DDR. DDR3, a more recent update, obtains eight datawords per access. 15