Upload presentasi
Presentasi sedang didownload. Silahkan tunggu
Diterbitkan olehFauzan Fauzy Telah diubah "9 tahun yang lalu
1
IKI10230 Pengantar Organisasi Komputer Bab 5.1: Memori
Sumber: 1. Hamacher. Computer Organization, ed Materi kuliah CS152/1997, UCB. 9 April 2003 Bobby Nazief Qonita Shahab bahan kuliah:
2
Memori: Tempat Penyimpanan Data
Keyboard, Mouse Computer Processor (active) Memory (passive) (where programs, data live when running) Devices Input Control (“brain”) Disk (permanent storages) Datapath (“brawn”) Output That is, any computer, no matter how primitive or advance, can be divided into five parts: 1. The input devices bring the data from the outside world into the computer. 2. These data are kept in the computer’s memory until ... 3. The datapath request and process them. 4. The operation of the datapath is controlled by the computer’s controller. All the work done by the computer will NOT do us any good unless we can get the data back to the outside world. 5. Getting the data back to the outside world is the job of the output devices. The most COMMON way to connect these 5 components together is to use a network of busses. Display, Printer
3
Istilah/Jenis Semikonduktor Memori
RAM --Random Access Memory time taken to access any arbitrary location in memory is constant SRAM --Static RAM A RAM chip design technology (see later) DRAM --Dynamic RAM A RAM chip design technology (see later) ROM --Read Only Memory ROMs are RAMs with data built-in when the chip is created. Usually stores BIOS info. Older uses included storage of bootstrap info PROM --Programmable ROM A ROM which can be programmed EPROM --Erasable PROM A PROM which can be programmed, erased by exposure to UV radiation EEROM – Electrical EPROM A PROM programmed & erased electrically
4
Masih tentang Istilah …
Tambahan istilah: SIMM Single In-Line Memory Module A packaging technology (single 32-bit data path) DIMM Dual In-Line Memory Module A packaging technology (dual 32-bit data paths) FPM RAM Fast Page-Mode RAM An older technology capable of about 60ns cycle time EDO RAM Extended-data-out RAM More modern FPM RAM, exploiting address coherency (see cache`later) capable of about 20ns access speed SDRAM Synchronous DRAM Synchronous Dynamic RAM; allows access speeds as low as about 10ns PC 100, PC133, PC2100, PC2600 => memory product you can buy
5
Connection: Memory - Processor
k-bit address bus Memory Sampai 2k addressable locations MAR n-bit data bus Panjang word = n bits MDR Control lines, R/W, MFC, etc.
6
Konsep Dasar Memory: akses per byte
Transfer dilakukan per-word (cepat, kelipatan bytes) Misalkan: 32-bit komputer => address 32 bit Kemampuan addressing: 2 ^ 32 = 4 Gbytes Jika transfer data per-word: 32 bit (data bus) => 4 bytes Bytes mana yang diakses dari kemungkinan word tsb? Perlu 2 bits untuk menentukan bytes yang mana dari word Sisa bit: 30 bits digunakan untuk address word
7
Organisasi Internal Memori
Bentuk array: terdiri dari sel memori Sel berisi 1 bit informasi Baris dari sel membentuk untaian satu word Contoh: 16 x 8 memori memori SRAM mengandung 16 words setiap words terdiri dari 8 bit data Kapasitas memori: 16 x 8 = 128 bits Decoder digunakan untuk memilih baris word mana yang akan diakses Tipikal SRAM, array 1 dimensi => indeks dari baris pada array tersebut.
8
Review: Static RAM Cell
6-Transistor SRAM Cell Latch menyimpan state 1 bit Transistor T bertindak sebagai switch Contoh: state 1 Latch dapat berubah dengan: put bit value pada b dan b’ word line pull high (select) word (row select) 1 T T 1 b’ b Write: 1. Drive bit lines sesuai dengan bit (mis. b = 1, b’ = 0) 2. Select row store nilai b dan b’ menjadi state latch Read: Precharge (set) bit lines high Select row 3. Sense amp mendeteksi bit lines mana yang low state bit The classical SRAM cell looks like this. It consists of two back-to-back inverters that serves as a flip-flop. Here is an expanded view of this cell, you can see it consists of 6 transistors. In order to write a value into this cell, you need to drive from both sides. For example, if you want to write a 1, you will drive “bit” to 1 while at the same time, drive “bit bar” to zero. Once the bit lines are driven to their desired values, you will turn on these two transistors by setting the word line to high so the values on the bit lines will be written into the cell. Remember now these are very very tiny transistors so we cannot rely on them to drive these long bit lines effectively during read. Also, the pull down devices are usually much stronger than the pull up devices. So the first thing we need to do on read is to charge these two bit lines to a high values. Once these bit lines are charged to high, we will turn on these two transistors so one of these inverters (the lower one in our example) will start pulling one of the bit line low while the other bit line will remain at HI. It will take this small inverter a long time to drive this long bit line to low but we don’t have to wait that long since all we need to detect the difference between these two bit lines. And if you ask any circuit designer, they will tell you it is much easier to detect a “differential signal” (point to bit and bit bar) than to detect an absolute signal. +2 = 30 min. (Y:10)
9
Organisasi Memori: 1-level-decode SRAM (128 x 8)
Word 8 bit data b7 b7’ b1 b1’ b0 b0’ Address decoder W0 A0 W1 A1 memory cells A6 W127 128 words sense/write amps sense/write amps sense/write amps R/W’ CS Input/output lines d7 d1 d0
10
Organisasi Memori: 2-level-decode SRAM (1 K x 1)
W0 5-bit decoder W1 32 x 32 Memory cell array W31 A0 5-bit row address A1 Sense/write circuitry A7 32 x 1 Output/input multiplexer R/W CS A8 A9 5-bit column address 10-bit address Data Input/Output (1 bit)
11
Static RAM (SRAM) SRAM dapat menyimpan “state” (isi RAM) selama terdapat “tegangan” power supply Sangat cepat, 10 nano-detik Densitas rendah (bits per chip) memerlukan 6 transistor per-sel mahal Pilihan teknologi untuk memori yang sangat cepat dengan kapasitas kecil cache
12
Review: 1-Transistor Memory Cell (DRAM)
Kapasitor menyimpan state 1 (charged) atau 0 (discharge) Perlu refresh! row select Write: 1. Drive bit line 2. Select row (T sebagai switch) Read: 1. Select row 2. Sense Amp (terhubung dengan bit line): sense & drives sesuai dengan value (threshold) 3. Write: restore the value (high or low) Refresh Just do a dummy read to every cell. T C bit The state of the art DRAM cell only has one transistor. The bit is stored in a tiny transistor. The write operation is very simple. Just drive the bit line and select the row by turning on this pass transistor. For read, we will need to precharge this bit line to high and then turn on the pass transistor. This will cause a small voltage change on the bit line and a very sensitive amplifier will be used to measure this small voltage change with respect to a reference bit line. Once again, the value we stored will be destroyed by the read operation so an automatic write back has to be performed at the end of every read. + 2 = 48 min. (Y:28)
13
Classical DRAM Organization (square)
bit (data) lines r o w d e c Each intersection represents a 1-T DRAM Cell RAM Cell Array word (row) select Similar to SRAM, DRAM is organized into rows and columns. But unlike SRAM, which allows you to read an entire row out at a time at a word, classical DRAM only allows you read out one-bit at time time. The reason for this is to save power as well as area. Remember now the DRAM cell is very small we have a lot of them across horizontally. So it will be very difficult to build a Sense Amplifier for each column due to the area constraint not to mention having a sense amplifier per column will consume a lot of power. You select the bit you want to read or write by supplying a Row and then a Column address. Similar to SRAM, each row control line is referred to as the word line and each vertical data line is referred to as the bit line. +2 = 57 min. (Y:37) Column Selector & I/O Circuits row address Column Address Row and Column Address together: Select 1 bit a time data
14
Densitas tinggi: 1 transistor/bit
Dynamic RAM (DRAM) Slower than SRAM access time ~60 ns (paling cepat: 35 ns) Nonpersistant every row must be accessed every ~1 ms (refreshed) Densitas tinggi: 1 transistor/bit Lebih murah dari SRAM ~$1/MByte [2002] Fragile electrical noise, light, radiation Pilihan teknologi memori untuk kapasitas besar dan “low cost” main memory
15
Organisasi DRAM 2-level (64Kx1)
RAS’ 256 Rows Row decoder 256x256 cell array Row address latch \ 8 row 256 Columns A15-A8/ A7-A0 column sense/write amps CS R/W’ col Column address latch column decoder \ 8 CAS’ Dout Din
16
Rewrite/Refreshed (~30ns)
Operasi DRAM Row Address (~50ns) Set Row address pada address lines & strobe RAS Seluruh row dibaca & disimpan di column latches Isi dari row memori cells akan di-refresh Column Address (~10ns) Set Column address pada address lines & strobe CAS Access selected bit READ: transfer from selected column latch to Dout WRITE: Set selected column latch to Din Rewrite/Refreshed (~30ns) Write back entire row
17
DRAM Write Timing RAS’ CAS’ R/W’ A 256K x 8 DRAM D DRAM WR Cycle Time
9 8 DRAM WR Cycle Time RAS’ CAS’ A Row Address Col Address Junk Row Address Col Address Junk Let me show you an example. Here we are performing two write operation to the DRAM. Setup/hold times in colar bars Remember, this is very important. All DRAM access start with the assertion of the RAS line. When the RAS_L line go low, the address lines are latched in as row address. This is followed by the CAS_L line going low to latch in the column address. Of course, there will be certain setup and hold time requirements for the address as well as data as highlighted here. Since the Write Enable line is already asserted before CAS is asserted, write will occur shortly after the column address is latched in. This is referred to as the Early Write Cycle. This is different from the 2nd example I showed here where the Write Enable signal comes AFTER the assertion of CAS. This is referred to as a Later Write cycle. Notice that in the early write cycle, the width of the CAS line, which you as a logic designer can and should control, must be as long as the memory’s write access time. On the other hand, in the later write cycle, the width of the Write Enable pulse must be as wide as the WR Access Time. Also notice that the RAS line has to remain asserted (low) during the entire access cycle. The DRAM write cycle time is defined as the time between the two RAS pulse and is much longer than the DRAM write access time. +3 = 63 min. (Y:43) R/W’ D Junk Data In Junk Data In Junk WR Access Time WR Access Time
18
DRAM Read Timing RAS’ CAS’ R/W’ A 256K x 8 DRAM D DRAM Read Cycle Time
9 8 DRAM Read Cycle Time RAS’ CAS’ A Row Address Col Address Junk Row Address Col Address Junk Similar to DRAM write, DRAM read can also be a Early read or a Late read. In the Early Read Cycle, Output Enable is asserted before CAS is asserted so the data lines will contain valid data one Read access time after the CAS line has gone low. In the Late Read cycle, Output Enable is asserted after CAS is asserted so the data will not be available on the data lines until one read access time after OE is asserted. Once again, notice that the RAS line has to remain asserted during the entire time. The DRAM read cycle time is defined as the time between the two RAS pulse. Notice that the DRAM read cycle time is much longer than the read access time. Q: RAS & CAS at same time? Yes, both must be low +2 = 65 min. (Y:45) R/W’ D High Z Junk Data Out High Z Data Out Read Access Time Read Access Time
19
Must Refresh Periodically
DRAM: Kinerja Timing Access time = 60ns < cycle time = 90ns Need to rewrite row Model asinkron: operasi memori dilakukan oleh controller circuit delay prosesor menunggu sampai cycle time selesai lalu melakukan request lagi. Must Refresh Periodically Perform complete memory cycle for each row Approx. every 1ms Handled in background by memory controller
20
Perkembangan Teknologi Memori DRAM
Teknologi memori: segi kecepatan akses berkembang sangat lambat Gap yang semakin membesar dengan kecepatan prosesor (cycle sangat kecil => 1 nsec, akses memori orde puluhan nsec). Perkembangan teknologi DRAM Basis tetap sama: 1-transistor memori cell (menggunakan kapasitor) Inovasi dilakukan dari segi: cara melakukan akses memotong waktu akses (mis. CAS tidak diperlukan) burst mode: sekaligus mengambil data sebanyak mungkin (seluruh word) perlu tambahan rangkaian: register, latch dll
21
Enhanced Performance DRAMs
Conventional Access Row + Col RAS CAS RAS CAS ... Page Mode Row + Series of columns RAS CAS CAS CAS ... Gives successive bits Video RAM Shift out entire row sequentially At video rate Row address latch Column decoder 256x256 cell array sense/write amps column & latch A15-A8/ A7-A0 \ 8 R/W’ CAS RAS row col Entire row buffered here Typical Performance row access time col access time cycle time page mode cycle time 50ns ns 90ns ns
22
Fast Page Mode Operation
Column Address Fast Page Mode DRAM N x M “SRAM” to save a row After a row is read into the register Only CAS is needed to access other M-bit blocks on that row RAS’ remains asserted while CAS’ is toggled N cols DRAM Row Address N rows N x M “SRAM” M bits M-bit Output So with this register in place, all we need to do is assert the RAS to latch in the row address, then entire row is read out and save into this register. After that, you only need to provide the column address and assert the CAS needs to access other M-bit within this same row. I like to point out that even I use the word “SRAM” here but this is no ordinary sram. It has to be very small but the good thing is that it is internal to the DRAM and does not have to drive any external load. Anyway, this type of operation where RAS remains asserted while CAS is toggled to bring in a new column address is called Page Mode operation. Strore orw so don’t have to repeat: SRAM It will become clearer why this is called Page Mode operation when we look into the operation of the SPARCstation 20 memory system. + 2 = 71 min. (Y:51) 1st M-bit Access 2nd M-bit 3rd M-bit 4th M-bit RAS’ CAS’ A Row Address Col Address Col Address Col Address Col Address
23
SDRAM & DDR SDRAM SDRAM: Synchronous DRAM
Address & Data are buffered in registers Burst Mode: Read/Write of different data lengths CAS signals are provided internally Standards: PC100, PC133 DDR SDRAM: Double-Data-Rate SDRAM Data is transferred on both edges of the clock Cell array is organized in 2 banks allows interleaving of word’s access Standards: PC2100, PC2300 RDRAM: Rambus DRAM High transfer rate using differential signaling Memory cells are organized in multiple banks Standards: proprietary owned by Rambus Inc.
24
SDRAM Operation Memory Latency: Memory Bandwidth:
RAS’ CAS’ Addr Row Col Data D0 D1 D2 D3 So with this register in place, all we need to do is assert the RAS to latch in the row address, then entire row is read out and save into this register. After that, you only need to provide the column address and assert the CAS needs to access other M-bit within this same row. I like to point out that even I use the word “SRAM” here but this is no ordinary sram. It has to be very small but the good thing is that it is internal to the DRAM and does not have to drive any external load. Anyway, this type of operation where RAS remains asserted while CAS is toggled to bring in a new column address is called Page Mode operation. Strore orw so don’t have to repeat: SRAM It will become clearer why this is called Page Mode operation when we look into the operation of the SPARCstation 20 memory system. + 2 = 71 min. (Y:51) Memory Latency: Waktu yang dibutuhkan untuk mentransfer word pertama Memory Bandwidth: Jumlah word (byte/bit) yang dapat ditransfer per satuan waktu
25
Struktur Memori Besar (1/4)
Misalkan: Chip memori 128K x 8 8 data lines 17 address lines CS’ Chips select WE’ CS’ WE’ Function Data Lines H X not selected Hi-Z L H Read data at location on address lines L L Write write data on data lines to address on address lines
26
Contoh: Struktur 1 MB (2/4)
1 MB dapat dikonstruksi dengan organisasi 8 chips memori 128 KB (8 x 128 x 8 = 1 MB) The address space is partitioned into 128K blocks; block 0 has addresses K -1 block 1 has addresses 128K K-1 block 2 has addresses 256K K -1 : block 7 has addresses 896K K -1 This will be chip 0 This will be chip 1 This will be chip 7 Berapa banyak bits yang diperlukan untuk alamat pada chips? memilih chips yang mana?
27
Contoh: Pembagian field address (3/4)
1MB membutuhkan alamat sebesar 20 bit, Ide: membagi field address menjadi 2 yakni: bits untuk memilih chips dan address pada field tsb. Bits (3 address bits) Bits (17 address bits) 17 bits select the address in each 128KB block (== each chip) 3 address bits select on of the 8 128KB blocks (chips)
28
Contoh: Struktur 1MB memory (4/4)
decoder 20 Address Lines 17 Address Lines A16 -- A0 3 Address Lines A19--A17 Write /WE Data Lines
29
Read-Only Memory ROM PROM EPROM EEPROM Flash
Write once, by manufacturer PROM Write once, by user EPROM Erasable PROM (by exposing it to ultraviolet light) EEPROM Electrically, Erasable PROM Flash ~EEPROM Write in blocks Low power consumption battery driven Implementation: Flash Cards Flash Drives: Better than disk (no movable parts faster response)
30
Ringkasan (.. To remember)
DRAM lambat tapi murah dan kapasitas besar (densitas tinggi) Pilihan untuk memberikan kapasitas BESAR pada sistem memori. SRAM cepat tapi mahal dan kapasitas kecil Pilihan untuk menyediakan sistem memori yang waktu aksesnya CEPAT. Struktur memori besar dapat dibangun dari kumpulan chips memori kecil: Field alamat dibagi: field address dan field untuk memilih chips/memori yang mana. Next topic: Trend teknologi memori (go to: http// search SDRAM guide)
31
Trend Teknologi Memori (DRAM)
CPU-DRAM Gap µProc 60%/yr. DRAM 7%/yr. 1 10 100 1000 1980 1981 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 CPU 1982 Processor-Memory Performance Gap: (grows 50% / year) Performance “Moore’s Law” Y-axis is performance X-axis is time Latency Cliché: Not e that x86 didn’t have cache on chip until 1989 Prosesor sangat cepat tidak efektif => kendala “bottleneck” berada pada sumber/tujuan data yakni memori.
Presentasi serupa
© 2024 SlidePlayer.info Inc.
All rights reserved.