SlideShare a Scribd company logo
1 of 32
Download to read offline
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM Circuit and
Architecture Basics
• Overview
• Terminology
• Access Protocol
• Architecture
Storage element
Switching
element
Bit Line
Word Line
(capacitor)
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM Circuit Basics
DRAM Cell
... Bit Lines...
Memory
Array
RowDecoder
...WordLines...
DRAM
Storage element
Switching
element
Bit Line
Word Line Data In/Out
Buffers
Sense Amps
Column Decoder
(capacitor)
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
Row, Bitlines and Wordlines
DRAM Circuit Basics
“Row” Defined
Bit Lines
Word Line
“Row” of DRAM
Row Size: 8 Kb @ 256 Mb SDRAM node
4 Kb @ 256 Mb RDRAM node
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM Circuit Basics
Sense Amplifier I
Sense
and
Amplify
6 Rows shown
1
2
3
4
5
6
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM Circuit Basics
Sense Amplifier II : Precharged
Vcc (logic 1) Gnd (logic 0) Vcc/2
precharged to Vcc/2
Sense
and
Amplify
1
2
3
4
5
6
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM Circuit Basics
Sense Amplifier III : Destructive Read
Sense
and
Amplify
1
2
3
4
5
6
Vcc (logic 1) Gnd (logic 0) Vcc/2
Wordline
Driven
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM Access Protocol
ROW ACCESS
AKA: OPEN a DRAM Page/Row
RAS (Row Address Strobe)
or
or
ACT (Activate a DRAM Page/Row)
BUS
MEMORY
CONTROLLERCPU
... Bit Lines...
Memory
Array
RowDecoder
...WordLines...
DRAM
Data In/Out
Buffers
Sense Amps
Column Decoder
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
once the data is valid on ALL of
the bit lines, you can select a
subset of the bits and send them
to the output buffers ... CAS
picks one of the bits
big point: cannot do another
RAS or precharge of the lines
until finished reading the column
data ... can’t change the values
on the bit lines or the output of
the sense amps until it has been
read by the memory controller
DRAM Circuit Basics
“Column” Defined
“One Row” of DRAM
Column: Smallest addressable quantity of DRAM on chip
SDRAM*: column size == chip data bus width (4, 8,16, 32)
RDRAM: column size != chip data bus width (128 bit fixed)
4 bit wide columns
SDRAM*: get “n” columns per access. n = (1, 2, 4, 8)
RDRAM: get 1 column per access.
#2 #3 #4 #5#0 #1
* SDRAM means SDRAM and variants. i.e. DDR SDRAM
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM Access Protocol
COLUMN ACCESS I
READ Command
or
CAS: Column Address Strobe
BUS
MEMORY
CONTROLLERCPU
... Bit Lines...
Memory
Array
RowDecoder
...WordLines...
DRAM
Data In/Out
Buffers
Sense Amps
Column Decoder
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
then the data is valid on the data
bus ... depending on what you
are using for in/out buffers, you
might be able to overlap a litttle
or a lot of the data transfer with
the next CAS to the same page
(this is PAGE MODE)
DRAM Access Protocol
Column Access II
note: page mode enables overlap with CAS
BUS
MEMORY
CONTROLLERCPU
... Bit Lines...
Memory
Array
RowDecoder
...WordLines...
DRAM
Data In/Out
Buffers
Sense Amps
Column Decoder
... with optional additional
CAS: Column Address Strobe
Data Out
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
NOTE
DRAM “Speed” Part I
How fast can I move data from DRAM cell to
sense amp?
RCD (Row Command Delay)
BUS
MEMORY
CONTROLLERCPU
... Bit Lines...
Memory
Array
RowDecoder
...WordLines...
DRAM
Data In/Out
Buffers
Sense Amps
Column Decoder
tRCD
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM “Speed” Part II
How fast can I get data out of sense amps
back into memory controller?
BUS
MEMORY
CONTROLLERCPU
... Bit Lines...
Memory
Array
RowDecoder
...WordLines...
DRAM
Data In/Out
Buffers
Sense Amps
Column Decoder
CAS: Column Address Strobe
tCAS aka
tCL
tCASL aka
CASL: Column Address Strobe Latency
CL: Column Address Strobe Latency
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM “Speed” Part III
How fast can I move data from DRAM cell into
memory controller?
RAC (Random Access Delay)
BUS
MEMORY
CONTROLLERCPU
... Bit Lines...
Memory
Array
RowDecoder
...WordLines...
DRAM
Data In/Out
Buffers
Sense Amps
Column Decoder
tRAC = tRCD + tCAS
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM “Speed” Part IV
How fast can I precharge DRAM array so I can
engage another RAS?
RP (Row Precharge Delay)
BUS
MEMORY
CONTROLLERCPU
... Bit Lines...
Memory
Array
RowDecoder
...WordLines...
DRAM
Data In/Out
Buffers
Sense Amps
Column Decoder
tRP
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM “Speed” Part V
How fast can I read from different rows?
RC (Row Cycle Time)
BUS
MEMORY
CONTROLLERCPU
... Bit Lines...
Memory
Array
RowDecoder
...WordLines...
DRAM
Data In/Out
Buffers
Sense Amps
Column Decoder
tRC = tRAS + tRP
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM “Speed” Summary I
What do I care about?
RC :Row Cycle Time
tRC = tRAS + tRP
RP :Row Precharge Delay
tRP
RAC :Random Access Delay
tRAC = tRCD + tCAS
CAS: Column Address Strobe
tCAS
RAS: Row Address Strobe
tRCD Seen in ads.
Easy to explain
Embedded systems designers
DRAM manufactuers
Computer Architect:
Latency bound code
i.e. linked list traversal
Easy to sell
RCD: Row Command Delay
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM “Speed” Summary II
DRAM Type Frequency
Data Bus
Width
(per chip)
Peak Data
Bandwidth
(per Chip)
Random
Access
Time (tRAC)
Row Cycle
Time (tRC)
PC133
SDRAM
133 16 200 MB/s 45 ns 60 ns
DDR 266 133 * 2 16 532 MB/s 45 ns 60 ns
PC800
RDRAM
400 * 2 16 1.6 GB/s 60 ns 70 ns
FCRAM 200 * 2 16 0.8 GB/s 25 ns 25 ns
RLDRAM 300 * 2 32 2.4 GB/s 25 ns 25 ns
DRAM is “slow”
But doesn’t have to be
tRC < 10ns achievable
Higher die cost
Not commodity Expensive
Not adopted in standard
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
DRAM “latency” isn’t
deterministic because of CAS or
RAS+CAS, and there may be
significant queuing delays within
the CPU and the memory
controller
Each transaction has some
overhead. Some types of
overhead cannot be pipelined.
This means that in general,
longer bursts are more efficient.
“DRAM latency”
B
C
D
DRAM
E2/E3
E1
F
A
CPU Mem
Controller
A: Transaction request may be delayed in Queue
B: Transaction request sent to Memory Controller
C: Transaction converted to Command Sequences
(may be queued)
D: Command/s Sent to DRAM
E1: Requires only a CAS or
E2: Requires RAS + CAS or
F: Transaction sent back to CPU
“DRAM Latency” = A + B + C + D + E + F
E3: Requires PRE + RAS + CAS
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
NOTE
DRAM Architecture Basics
PHYSICAL ORGANIZATION
This is per bank …
Typical DRAMs have 2+ banks
x2 DRAM x4 DRAM x8 DRAM
... Bit Lines...
Sense Amps
....
Data
Buffers
x8 DRAM
RowDecoder
Memory
Array
Column Decoder
... Bit Lines...
Sense Amps
....Data
Buffers
x2 DRAM
RowDecoder
Memory
Array
Column Decoder
... Bit Lines...
Sense Amps
....
Data
Buffers
x4 DRAM
RowDecoder
Memory
Array
Column Decoder
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
let’s look at the interface another
way .. the say the data sheets
portray it.
[explain]
main point: the RAS and CAS
signals directly control the
latches that hold the row and
column addresses ...
DRAM Architecture Basics
Read Timing for Conventional DRAM
Row
Address
Column
Address
Valid
Dataout
RAS
CAS
Address
DQ
Row
Address
Column
Address
Valid
Dataout
Data Transfer
Column Access
Row Access
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
since DRAM’s inception, there
have been a stream of changes
to the design, from FPM to EDO
to Burst EDO to SDRAM. the
changes are largely structural
modifications -- nimor -- that
target THROUGHPUT.
[discuss FPM up to SDRAM
Everything up to and including
SDRAM has been relatively
inexpensive, especially when
considering the pay-off (FPM
was essentially free, EDO cost a
latch, PBEDO cost a counter,
SDRAM cost a slight re-design).
however, we’re run out of “free”
ideas, and now all changes are
considered expensive ... thus
there is no consensus on new
directions and myriad of choices
has appeared
[ do LATENCY mods starting
with ESDRAM ... and then the
INTERFACE mods ]
DRAM Evolutionary Tree
(Mostly) Structural Modifications
Interface Modifications
Structural
Conventional
FPM EDO ESDRAM
Rambus, DDR/2 Future Trends
........
. . . . . .
MOSYS
FCRAM
VCDRAM
$
Modifications
Targeting
Latency
Targeting Throughput
Targeting Throughput
DRAM
SDRAMP/BEDO
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
NOTE
DRAM Evolution
Read Timing for Conventional DRAM
Row
Address
Column
Address
Valid
Dataout
RAS
CAS
Address
DQ
Row
Address
Column
Address
Valid
Dataout
Data Transfer
Column Access
Transfer Overlap
Row Access
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
FPM aallows you to keep th
esense amps actuve for multiple
CAS commands ...
much better throughput
problem: cannot latch a new
value in the column address
buffer until the read-out of the
data is complete
DRAM Evolution
Read Timing for Fast Page Mode
Row
Address
Column
Address
Valid
Dataout
Column
Address
Column
Address
Valid
Dataout
Valid
Dataout
RAS
CAS
Address
DQ
Data Transfer
Column Access
Transfer Overlap
Row Access
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
solution to that problem --
instead of simple tri-state
buffers, use a latch as well.
by putting a latch after the
column mux, the next column
address command can begin
sooner
DRAM Evolution
Read Timing for Extended Data Out
Row
Address
Column
Address
Valid
Dataout
RAS
CAS
Address
DQ
Column
Address
Column
Address
Valid
Dataout
Valid
Dataout
Data Transfer
Column Access
Transfer Overlap
Row Access
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
by driving the col-addr latch from
an internal counter rather than
an external signal, the minimum
cycle time for driving the output
bus was reduced by roughly
30%
DRAM Evolution
Read Timing for Burst EDO
Row
Address
Column
Address
RAS
CAS
Address
DQ
Data Transfer
Column Access
Transfer Overlap
Row Access
Valid
Data
Valid
Data
Valid
Data
Valid
Data
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
“pipeline” refers to the setting up
of the read pipeline ... first CAS
toggle latches the column
address, all following CAS
toggles drive data out onto the
bus. therefore data stops coming
when the memory controller
stops toggling CAS
DRAM Evolution
Read Timing for Pipeline Burst EDO
Row
Address
Column
Address
RAS
CAS
Address
DQ
Data Transfer
Column Access
Transfer Overlap
Row Access
Valid
Data
Valid
Data
Valid
Data
Valid
Data
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
main benefit: frees up the CPU
or memory controller from
having to control the DRAM’s
internal latches directly ... the
controller/CPU can go off and do
other things during the idle
cycles instead of “wait” ... even
though the time-to-first-word
latency actually gets worse, the
scheme increases system
throughput
DRAM Evolution
Read Timing for Synchronous DRAM
(RAS + CAS + OE ... == Command Bus)
Command
Address
DQ
Clock
Row
Addr
Col
Addr
Valid
Data
Valid
Data
Valid
Data
Valid
Data
ACT READ
RAS
CAS
Data Transfer
Column Access
Transfer Overlap
Row Access
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
output latch on EDO allowed you
to start CAS sooner for next
accesss (to same row)
latch whole row in ESDRAM --
allows you to start precharge &
RAS sooner for thee next page
access -- HIDE THE
PRECHARGE OVERHEAD.
DRAM Evolution
Inter-Row Read Timing for ESDRAM
Command
Address
DQ
Clock
Row
Addr
Col
Addr
Valid
Data
Valid
Data
Valid
Data
Valid
Data
ACT READ
Row
Addr
Col
Addr
Valid
Data
Valid
Data
Valid
Data
Valid
Data
ACT READPRE
“Regular” CAS-2 SDRAM, R/R to same bank
Command
Address
DQ
Clock
Row
Addr
Col
Addr
Valid
Data
Valid
Data
Valid
Data
Valid
Data
ACT READ
Row
Addr
Col
Addr
Valid
Data
Valid
Data
Valid
Data
Valid
Data
ACT READ
ESDRAM, R/R to same bank
PRE
Bank
Bank
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
neat feature of this type of
buffering: write-around
DRAM Evolution
Write-Around in ESDRAM
(can second READ be this aggressive?)
Command
Address
DQ
Clock
Row
Addr
Col
Addr
Valid
Data
Valid
Data
Valid
Data
Valid
Data
ACT READ
Row
Addr
Col
Addr
Valid
Data
Valid
Data
Valid
Data
Valid
Data
ACT WRITEPRE
“Regular” CAS-2 SDRAM, R/W/R to same bank, rows 0/1/0
Command
Address
DQ
Clock
Row
Addr
Col
Addr
Valid
Data
Valid
Data
Valid
Data
Valid
Data
ACT READ
Row
Addr
Col
Addr
Valid
Data
Valid
Data
Valid
Data
Valid
Data
ACT WRITE
ESDRAM, R/W/R to same bank, rows 0/1/0
PRE
Bank
Bank
Row
Addr
Col
Addr
Valid
Data
Valid
Data
Valid
Data
ACT READPRE
Bank
Col
Addr
Valid
Data
Valid
Data
Valid
Data
Valid
Data
READ
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
main thing ... it is like having a
bunch of open row buffers (a la
rambus), but the problem is that
you must deal with the cache
directly (move into and out of it),
not the DRAM banks ... adds an
extra couple of cycles of latency
... however, you get good
bandwidth if the data you want is
cache, and you can “prefetch”
into cache ahead of when you
want it ... originally targetted at
reducing latency, now that
SDRAM is CAS-2 and RCD-2,
this make sense only in a
throughput way
DRAM Evolution
Internal Structure of Virtual Channel
Segment cache is software-managed, reduces energy
$
Row Decoder
2Kb Segment
2Kb Segment
2Kb Segment
2Kb Segment
Bank A
Bank B
16 “Channels”
Input/Output
Buffer
DQs
Sel/Dec
(segments)
Sense
Amps
2Kbit # DQs
Activate Prefetch
Restore
Read
Write
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
FCRAM opts to break up the
data array .. only activate a
portion of the word line
8K rows requires 13 bits tto
select ... FCRAM uses 15
(assuming the array is 8k x 1k ...
the data sheet does not specify)
DRAM Evolution
Internal Structure of Fast Cycle RAM
Reduces access time and energy/access
tRCD = 15ns tRCD = 5ns
8M Array
13 bits
Sense Amps
8M Array
15 bits
Sense Amps
SDRAM FCRAM
(one clock)(two clocks)
RowDecoder
RowDecoder
(8Kr x 1Kb) (?)
DRAM Memory
System: Lecture 2
Spring 2003
Bruce Jacob
David Wang
University of
Maryland
MoSys takes this one step
further ... DRAM with an SRAM
interface & speed but DRAM
energy
[physical partitioning: 72 banks]
auto refresh -- how to do this
transparently? the logic moves
tthrough the arrays, refreshing
them when not active.
but what is one bank gets
repeated access for a long
duration? all other banks will be
refreshed, but that one will not.
solution: they have a bank-sized
CACHE of lines ... in theory,
should never have a problem
(magic)
DRAM Evolution
Internal Structure of MoSys 1T-SRAM
........
. . . . . .
addr
Bank
Select
DQs
$
Auto
Refresh

More Related Content

What's hot

CPU - Central processing unit
CPU - Central processing unitCPU - Central processing unit
CPU - Central processing unitAbhinav Vishnoi
 
Memory Hierarchy (RAM and ROM)
Memory Hierarchy (RAM and ROM)Memory Hierarchy (RAM and ROM)
Memory Hierarchy (RAM and ROM)sumanth ch
 
Processors used in System on chip
Processors used in System on chip Processors used in System on chip
Processors used in System on chip A B Shinde
 
OMI - The Missing Piece of a Modular, Flexible and Composable Computing World
OMI - The Missing Piece of a Modular, Flexible and Composable Computing WorldOMI - The Missing Piece of a Modular, Flexible and Composable Computing World
OMI - The Missing Piece of a Modular, Flexible and Composable Computing WorldAllan Cantle
 
motherboard electronic components and their functions
motherboard electronic components and their functionsmotherboard electronic components and their functions
motherboard electronic components and their functionsOmprakash Chauhan
 
CPUs, Form Factor And Sockets
CPUs, Form Factor And SocketsCPUs, Form Factor And Sockets
CPUs, Form Factor And Socketsricsanmae
 
System On Chip (SOC)
System On Chip (SOC)System On Chip (SOC)
System On Chip (SOC)Shivam Gupta
 
08. Central Processing Unit (CPU)
08. Central Processing Unit (CPU)08. Central Processing Unit (CPU)
08. Central Processing Unit (CPU)Akhila Dakshina
 
Introduction to-microprocessor
Introduction to-microprocessorIntroduction to-microprocessor
Introduction to-microprocessorankitnav1
 

What's hot (20)

Cache memory
Cache memoryCache memory
Cache memory
 
CPU - Central processing unit
CPU - Central processing unitCPU - Central processing unit
CPU - Central processing unit
 
Memory Hierarchy (RAM and ROM)
Memory Hierarchy (RAM and ROM)Memory Hierarchy (RAM and ROM)
Memory Hierarchy (RAM and ROM)
 
Processors used in System on chip
Processors used in System on chip Processors used in System on chip
Processors used in System on chip
 
DDR
DDRDDR
DDR
 
OMI - The Missing Piece of a Modular, Flexible and Composable Computing World
OMI - The Missing Piece of a Modular, Flexible and Composable Computing WorldOMI - The Missing Piece of a Modular, Flexible and Composable Computing World
OMI - The Missing Piece of a Modular, Flexible and Composable Computing World
 
SRAM Design
SRAM DesignSRAM Design
SRAM Design
 
Motherboard ppt
Motherboard pptMotherboard ppt
Motherboard ppt
 
motherboard electronic components and their functions
motherboard electronic components and their functionsmotherboard electronic components and their functions
motherboard electronic components and their functions
 
Memory hierarchy
Memory hierarchyMemory hierarchy
Memory hierarchy
 
Dram and its types
Dram and its typesDram and its types
Dram and its types
 
Sram pdf
Sram pdfSram pdf
Sram pdf
 
Direct memory access
Direct memory accessDirect memory access
Direct memory access
 
Types of RAM
Types of RAMTypes of RAM
Types of RAM
 
CPUs, Form Factor And Sockets
CPUs, Form Factor And SocketsCPUs, Form Factor And Sockets
CPUs, Form Factor And Sockets
 
System On Chip (SOC)
System On Chip (SOC)System On Chip (SOC)
System On Chip (SOC)
 
08. Central Processing Unit (CPU)
08. Central Processing Unit (CPU)08. Central Processing Unit (CPU)
08. Central Processing Unit (CPU)
 
Introduction to-microprocessor
Introduction to-microprocessorIntroduction to-microprocessor
Introduction to-microprocessor
 
Microprocessor 8086
Microprocessor 8086Microprocessor 8086
Microprocessor 8086
 
How the cpu works
How the cpu worksHow the cpu works
How the cpu works
 

Viewers also liked

A performance Comparison of contemporary DRAM Architectures
A performance Comparison of contemporary  DRAM ArchitecturesA performance Comparison of contemporary  DRAM Architectures
A performance Comparison of contemporary DRAM ArchitecturesJuan Ribeiro Reis, MSc.
 
GDDR Solution Design and Implementation Techniques
GDDR Solution Design and Implementation Techniques GDDR Solution Design and Implementation Techniques
GDDR Solution Design and Implementation Techniques EMC
 
Memory mapping plan
Memory mapping planMemory mapping plan
Memory mapping planJames Chang
 
Embedded systems in biomedical applications
Embedded systems in biomedical applicationsEmbedded systems in biomedical applications
Embedded systems in biomedical applicationsSeminar Links
 

Viewers also liked (8)

A performance Comparison of contemporary DRAM Architectures
A performance Comparison of contemporary  DRAM ArchitecturesA performance Comparison of contemporary  DRAM Architectures
A performance Comparison of contemporary DRAM Architectures
 
GDDR Solution Design and Implementation Techniques
GDDR Solution Design and Implementation Techniques GDDR Solution Design and Implementation Techniques
GDDR Solution Design and Implementation Techniques
 
Memory mapping plan
Memory mapping planMemory mapping plan
Memory mapping plan
 
Embedded dram
Embedded dramEmbedded dram
Embedded dram
 
Esmeralda gonzalez de britton 2009
Esmeralda gonzalez de britton 2009Esmeralda gonzalez de britton 2009
Esmeralda gonzalez de britton 2009
 
Dynamic RAM
Dynamic RAMDynamic RAM
Dynamic RAM
 
DRAM
DRAMDRAM
DRAM
 
Embedded systems in biomedical applications
Embedded systems in biomedical applicationsEmbedded systems in biomedical applications
Embedded systems in biomedical applications
 

Similar to Lecture2

Microelectronics U4.pptx.ppt
Microelectronics U4.pptx.pptMicroelectronics U4.pptx.ppt
Microelectronics U4.pptx.pptPavikaSharma3
 
Lec11 Computer Architecture by Hsien-Hsin Sean Lee Georgia Tech -- Memory part3
Lec11 Computer Architecture by Hsien-Hsin Sean Lee Georgia Tech -- Memory part3Lec11 Computer Architecture by Hsien-Hsin Sean Lee Georgia Tech -- Memory part3
Lec11 Computer Architecture by Hsien-Hsin Sean Lee Georgia Tech -- Memory part3Hsien-Hsin Sean Lee, Ph.D.
 
IRJET- Design And VLSI Verification of DDR SDRAM Controller Using VHDL
IRJET- Design And VLSI Verification of DDR SDRAM Controller Using VHDLIRJET- Design And VLSI Verification of DDR SDRAM Controller Using VHDL
IRJET- Design And VLSI Verification of DDR SDRAM Controller Using VHDLIRJET Journal
 
Digital electronics and logic design
Digital electronics and logic designDigital electronics and logic design
Digital electronics and logic designToufiq Akbar
 
memeoryorganization PPT for organization of memories
memeoryorganization PPT for organization of memoriesmemeoryorganization PPT for organization of memories
memeoryorganization PPT for organization of memoriesGauravDaware2
 
301378156 design-of-sram-in-verilog
301378156 design-of-sram-in-verilog301378156 design-of-sram-in-verilog
301378156 design-of-sram-in-verilogSrinivas Naidu
 
Chapter5 the memory-system-jntuworld
Chapter5 the memory-system-jntuworldChapter5 the memory-system-jntuworld
Chapter5 the memory-system-jntuworldPraveen Kumar
 
Kiến trúc máy tính - COE 301 - Memory.ppt
Kiến trúc máy tính - COE 301 - Memory.pptKiến trúc máy tính - COE 301 - Memory.ppt
Kiến trúc máy tính - COE 301 - Memory.pptTriTrang4
 
Time and Low Power Operation Using Embedded Dram to Gain Cell Data Retention
Time and Low Power Operation Using Embedded Dram to Gain Cell Data RetentionTime and Low Power Operation Using Embedded Dram to Gain Cell Data Retention
Time and Low Power Operation Using Embedded Dram to Gain Cell Data RetentionIJMTST Journal
 
COMPUTER ORGANIZATION NOTES Unit 5
COMPUTER ORGANIZATION NOTES Unit 5COMPUTER ORGANIZATION NOTES Unit 5
COMPUTER ORGANIZATION NOTES Unit 5Dr.MAYA NAYAK
 

Similar to Lecture2 (20)

Microelectronics U4.pptx.ppt
Microelectronics U4.pptx.pptMicroelectronics U4.pptx.ppt
Microelectronics U4.pptx.ppt
 
Lec11 Computer Architecture by Hsien-Hsin Sean Lee Georgia Tech -- Memory part3
Lec11 Computer Architecture by Hsien-Hsin Sean Lee Georgia Tech -- Memory part3Lec11 Computer Architecture by Hsien-Hsin Sean Lee Georgia Tech -- Memory part3
Lec11 Computer Architecture by Hsien-Hsin Sean Lee Georgia Tech -- Memory part3
 
IRJET- Design And VLSI Verification of DDR SDRAM Controller Using VHDL
IRJET- Design And VLSI Verification of DDR SDRAM Controller Using VHDLIRJET- Design And VLSI Verification of DDR SDRAM Controller Using VHDL
IRJET- Design And VLSI Verification of DDR SDRAM Controller Using VHDL
 
Memory management
Memory managementMemory management
Memory management
 
DDR DIMM Design
DDR DIMM DesignDDR DIMM Design
DDR DIMM Design
 
Digital electronics and logic design
Digital electronics and logic designDigital electronics and logic design
Digital electronics and logic design
 
Memory (Part 3)
Memory (Part 3)Memory (Part 3)
Memory (Part 3)
 
Memory technologies
Memory technologiesMemory technologies
Memory technologies
 
Presentation1
Presentation1Presentation1
Presentation1
 
Memoryhierarchy
MemoryhierarchyMemoryhierarchy
Memoryhierarchy
 
memeoryorganization PPT for organization of memories
memeoryorganization PPT for organization of memoriesmemeoryorganization PPT for organization of memories
memeoryorganization PPT for organization of memories
 
301378156 design-of-sram-in-verilog
301378156 design-of-sram-in-verilog301378156 design-of-sram-in-verilog
301378156 design-of-sram-in-verilog
 
Chapter5 the memory-system-jntuworld
Chapter5 the memory-system-jntuworldChapter5 the memory-system-jntuworld
Chapter5 the memory-system-jntuworld
 
Kiến trúc máy tính - COE 301 - Memory.ppt
Kiến trúc máy tính - COE 301 - Memory.pptKiến trúc máy tính - COE 301 - Memory.ppt
Kiến trúc máy tính - COE 301 - Memory.ppt
 
05 Internal Memory
05  Internal  Memory05  Internal  Memory
05 Internal Memory
 
Time and Low Power Operation Using Embedded Dram to Gain Cell Data Retention
Time and Low Power Operation Using Embedded Dram to Gain Cell Data RetentionTime and Low Power Operation Using Embedded Dram to Gain Cell Data Retention
Time and Low Power Operation Using Embedded Dram to Gain Cell Data Retention
 
COMPUTER ORGANIZATION NOTES Unit 5
COMPUTER ORGANIZATION NOTES Unit 5COMPUTER ORGANIZATION NOTES Unit 5
COMPUTER ORGANIZATION NOTES Unit 5
 
Module-5A.pdf
Module-5A.pdfModule-5A.pdf
Module-5A.pdf
 
Dd sdram
Dd sdramDd sdram
Dd sdram
 
Memory And Storages
Memory And StoragesMemory And Storages
Memory And Storages
 

More from philipsinter

More from philipsinter (10)

Fundamentals of Database system
Fundamentals of Database systemFundamentals of Database system
Fundamentals of Database system
 
MULTIMEDIA Cocomo forum version5
MULTIMEDIA Cocomo forum version5 MULTIMEDIA Cocomo forum version5
MULTIMEDIA Cocomo forum version5
 
Multimedia
Multimedia Multimedia
Multimedia
 
multimedia 01
multimedia 01multimedia 01
multimedia 01
 
Xml 215-presentation
Xml 215-presentationXml 215-presentation
Xml 215-presentation
 
Xml
XmlXml
Xml
 
Server side
Server sideServer side
Server side
 
Java servlet
Java servletJava servlet
Java servlet
 
Dbms
DbmsDbms
Dbms
 
Final vlsi projectreport
Final vlsi projectreportFinal vlsi projectreport
Final vlsi projectreport
 

Recently uploaded

Regression analysis: Simple Linear Regression Multiple Linear Regression
Regression analysis:  Simple Linear Regression Multiple Linear RegressionRegression analysis:  Simple Linear Regression Multiple Linear Regression
Regression analysis: Simple Linear Regression Multiple Linear RegressionRavindra Nath Shukla
 
VIP Call Girl Jamshedpur Aashi 8250192130 Independent Escort Service Jamshedpur
VIP Call Girl Jamshedpur Aashi 8250192130 Independent Escort Service JamshedpurVIP Call Girl Jamshedpur Aashi 8250192130 Independent Escort Service Jamshedpur
VIP Call Girl Jamshedpur Aashi 8250192130 Independent Escort Service JamshedpurSuhani Kapoor
 
Lucknow 💋 Escorts in Lucknow - 450+ Call Girl Cash Payment 8923113531 Neha Th...
Lucknow 💋 Escorts in Lucknow - 450+ Call Girl Cash Payment 8923113531 Neha Th...Lucknow 💋 Escorts in Lucknow - 450+ Call Girl Cash Payment 8923113531 Neha Th...
Lucknow 💋 Escorts in Lucknow - 450+ Call Girl Cash Payment 8923113531 Neha Th...anilsa9823
 
A DAY IN THE LIFE OF A SALESMAN / WOMAN
A DAY IN THE LIFE OF A  SALESMAN / WOMANA DAY IN THE LIFE OF A  SALESMAN / WOMAN
A DAY IN THE LIFE OF A SALESMAN / WOMANIlamathiKannappan
 
BEST Call Girls In Old Faridabad ✨ 9773824855 ✨ Escorts Service In Delhi Ncr,
BEST Call Girls In Old Faridabad ✨ 9773824855 ✨ Escorts Service In Delhi Ncr,BEST Call Girls In Old Faridabad ✨ 9773824855 ✨ Escorts Service In Delhi Ncr,
BEST Call Girls In Old Faridabad ✨ 9773824855 ✨ Escorts Service In Delhi Ncr,noida100girls
 
Tech Startup Growth Hacking 101 - Basics on Growth Marketing
Tech Startup Growth Hacking 101  - Basics on Growth MarketingTech Startup Growth Hacking 101  - Basics on Growth Marketing
Tech Startup Growth Hacking 101 - Basics on Growth MarketingShawn Pang
 
Insurers' journeys to build a mastery in the IoT usage
Insurers' journeys to build a mastery in the IoT usageInsurers' journeys to build a mastery in the IoT usage
Insurers' journeys to build a mastery in the IoT usageMatteo Carbone
 
Vip Female Escorts Noida 9711199171 Greater Noida Escorts Service
Vip Female Escorts Noida 9711199171 Greater Noida Escorts ServiceVip Female Escorts Noida 9711199171 Greater Noida Escorts Service
Vip Female Escorts Noida 9711199171 Greater Noida Escorts Serviceankitnayak356677
 
/:Call Girls In Jaypee Siddharth - 5 Star Hotel New Delhi ➥9990211544 Top Esc...
/:Call Girls In Jaypee Siddharth - 5 Star Hotel New Delhi ➥9990211544 Top Esc.../:Call Girls In Jaypee Siddharth - 5 Star Hotel New Delhi ➥9990211544 Top Esc...
/:Call Girls In Jaypee Siddharth - 5 Star Hotel New Delhi ➥9990211544 Top Esc...lizamodels9
 
Cash Payment 9602870969 Escort Service in Udaipur Call Girls
Cash Payment 9602870969 Escort Service in Udaipur Call GirlsCash Payment 9602870969 Escort Service in Udaipur Call Girls
Cash Payment 9602870969 Escort Service in Udaipur Call GirlsApsara Of India
 
Russian Faridabad Call Girls(Badarpur) : ☎ 8168257667, @4999
Russian Faridabad Call Girls(Badarpur) : ☎ 8168257667, @4999Russian Faridabad Call Girls(Badarpur) : ☎ 8168257667, @4999
Russian Faridabad Call Girls(Badarpur) : ☎ 8168257667, @4999Tina Ji
 
Grateful 7 speech thanking everyone that has helped.pdf
Grateful 7 speech thanking everyone that has helped.pdfGrateful 7 speech thanking everyone that has helped.pdf
Grateful 7 speech thanking everyone that has helped.pdfPaul Menig
 
The CMO Survey - Highlights and Insights Report - Spring 2024
The CMO Survey - Highlights and Insights Report - Spring 2024The CMO Survey - Highlights and Insights Report - Spring 2024
The CMO Survey - Highlights and Insights Report - Spring 2024christinemoorman
 
Intro to BCG's Carbon Emissions Benchmark_vF.pdf
Intro to BCG's Carbon Emissions Benchmark_vF.pdfIntro to BCG's Carbon Emissions Benchmark_vF.pdf
Intro to BCG's Carbon Emissions Benchmark_vF.pdfpollardmorgan
 
Call Girls in Gomti Nagar - 7388211116 - With room Service
Call Girls in Gomti Nagar - 7388211116  - With room ServiceCall Girls in Gomti Nagar - 7388211116  - With room Service
Call Girls in Gomti Nagar - 7388211116 - With room Servicediscovermytutordmt
 
Call Girls Pune Just Call 9907093804 Top Class Call Girl Service Available
Call Girls Pune Just Call 9907093804 Top Class Call Girl Service AvailableCall Girls Pune Just Call 9907093804 Top Class Call Girl Service Available
Call Girls Pune Just Call 9907093804 Top Class Call Girl Service AvailableDipal Arora
 
Call Girls In Connaught Place Delhi ❤️88604**77959_Russian 100% Genuine Escor...
Call Girls In Connaught Place Delhi ❤️88604**77959_Russian 100% Genuine Escor...Call Girls In Connaught Place Delhi ❤️88604**77959_Russian 100% Genuine Escor...
Call Girls In Connaught Place Delhi ❤️88604**77959_Russian 100% Genuine Escor...lizamodels9
 
RE Capital's Visionary Leadership under Newman Leech
RE Capital's Visionary Leadership under Newman LeechRE Capital's Visionary Leadership under Newman Leech
RE Capital's Visionary Leadership under Newman LeechNewman George Leech
 

Recently uploaded (20)

Regression analysis: Simple Linear Regression Multiple Linear Regression
Regression analysis:  Simple Linear Regression Multiple Linear RegressionRegression analysis:  Simple Linear Regression Multiple Linear Regression
Regression analysis: Simple Linear Regression Multiple Linear Regression
 
VIP Call Girl Jamshedpur Aashi 8250192130 Independent Escort Service Jamshedpur
VIP Call Girl Jamshedpur Aashi 8250192130 Independent Escort Service JamshedpurVIP Call Girl Jamshedpur Aashi 8250192130 Independent Escort Service Jamshedpur
VIP Call Girl Jamshedpur Aashi 8250192130 Independent Escort Service Jamshedpur
 
Lucknow 💋 Escorts in Lucknow - 450+ Call Girl Cash Payment 8923113531 Neha Th...
Lucknow 💋 Escorts in Lucknow - 450+ Call Girl Cash Payment 8923113531 Neha Th...Lucknow 💋 Escorts in Lucknow - 450+ Call Girl Cash Payment 8923113531 Neha Th...
Lucknow 💋 Escorts in Lucknow - 450+ Call Girl Cash Payment 8923113531 Neha Th...
 
A DAY IN THE LIFE OF A SALESMAN / WOMAN
A DAY IN THE LIFE OF A  SALESMAN / WOMANA DAY IN THE LIFE OF A  SALESMAN / WOMAN
A DAY IN THE LIFE OF A SALESMAN / WOMAN
 
BEST Call Girls In Old Faridabad ✨ 9773824855 ✨ Escorts Service In Delhi Ncr,
BEST Call Girls In Old Faridabad ✨ 9773824855 ✨ Escorts Service In Delhi Ncr,BEST Call Girls In Old Faridabad ✨ 9773824855 ✨ Escorts Service In Delhi Ncr,
BEST Call Girls In Old Faridabad ✨ 9773824855 ✨ Escorts Service In Delhi Ncr,
 
Tech Startup Growth Hacking 101 - Basics on Growth Marketing
Tech Startup Growth Hacking 101  - Basics on Growth MarketingTech Startup Growth Hacking 101  - Basics on Growth Marketing
Tech Startup Growth Hacking 101 - Basics on Growth Marketing
 
Insurers' journeys to build a mastery in the IoT usage
Insurers' journeys to build a mastery in the IoT usageInsurers' journeys to build a mastery in the IoT usage
Insurers' journeys to build a mastery in the IoT usage
 
Vip Female Escorts Noida 9711199171 Greater Noida Escorts Service
Vip Female Escorts Noida 9711199171 Greater Noida Escorts ServiceVip Female Escorts Noida 9711199171 Greater Noida Escorts Service
Vip Female Escorts Noida 9711199171 Greater Noida Escorts Service
 
/:Call Girls In Jaypee Siddharth - 5 Star Hotel New Delhi ➥9990211544 Top Esc...
/:Call Girls In Jaypee Siddharth - 5 Star Hotel New Delhi ➥9990211544 Top Esc.../:Call Girls In Jaypee Siddharth - 5 Star Hotel New Delhi ➥9990211544 Top Esc...
/:Call Girls In Jaypee Siddharth - 5 Star Hotel New Delhi ➥9990211544 Top Esc...
 
Cash Payment 9602870969 Escort Service in Udaipur Call Girls
Cash Payment 9602870969 Escort Service in Udaipur Call GirlsCash Payment 9602870969 Escort Service in Udaipur Call Girls
Cash Payment 9602870969 Escort Service in Udaipur Call Girls
 
Russian Faridabad Call Girls(Badarpur) : ☎ 8168257667, @4999
Russian Faridabad Call Girls(Badarpur) : ☎ 8168257667, @4999Russian Faridabad Call Girls(Badarpur) : ☎ 8168257667, @4999
Russian Faridabad Call Girls(Badarpur) : ☎ 8168257667, @4999
 
Grateful 7 speech thanking everyone that has helped.pdf
Grateful 7 speech thanking everyone that has helped.pdfGrateful 7 speech thanking everyone that has helped.pdf
Grateful 7 speech thanking everyone that has helped.pdf
 
KestrelPro Flyer Japan IT Week 2024 (English)
KestrelPro Flyer Japan IT Week 2024 (English)KestrelPro Flyer Japan IT Week 2024 (English)
KestrelPro Flyer Japan IT Week 2024 (English)
 
Best Practices for Implementing an External Recruiting Partnership
Best Practices for Implementing an External Recruiting PartnershipBest Practices for Implementing an External Recruiting Partnership
Best Practices for Implementing an External Recruiting Partnership
 
The CMO Survey - Highlights and Insights Report - Spring 2024
The CMO Survey - Highlights and Insights Report - Spring 2024The CMO Survey - Highlights and Insights Report - Spring 2024
The CMO Survey - Highlights and Insights Report - Spring 2024
 
Intro to BCG's Carbon Emissions Benchmark_vF.pdf
Intro to BCG's Carbon Emissions Benchmark_vF.pdfIntro to BCG's Carbon Emissions Benchmark_vF.pdf
Intro to BCG's Carbon Emissions Benchmark_vF.pdf
 
Call Girls in Gomti Nagar - 7388211116 - With room Service
Call Girls in Gomti Nagar - 7388211116  - With room ServiceCall Girls in Gomti Nagar - 7388211116  - With room Service
Call Girls in Gomti Nagar - 7388211116 - With room Service
 
Call Girls Pune Just Call 9907093804 Top Class Call Girl Service Available
Call Girls Pune Just Call 9907093804 Top Class Call Girl Service AvailableCall Girls Pune Just Call 9907093804 Top Class Call Girl Service Available
Call Girls Pune Just Call 9907093804 Top Class Call Girl Service Available
 
Call Girls In Connaught Place Delhi ❤️88604**77959_Russian 100% Genuine Escor...
Call Girls In Connaught Place Delhi ❤️88604**77959_Russian 100% Genuine Escor...Call Girls In Connaught Place Delhi ❤️88604**77959_Russian 100% Genuine Escor...
Call Girls In Connaught Place Delhi ❤️88604**77959_Russian 100% Genuine Escor...
 
RE Capital's Visionary Leadership under Newman Leech
RE Capital's Visionary Leadership under Newman LeechRE Capital's Visionary Leadership under Newman Leech
RE Capital's Visionary Leadership under Newman Leech
 

Lecture2

  • 1. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM Circuit and Architecture Basics • Overview • Terminology • Access Protocol • Architecture Storage element Switching element Bit Line Word Line (capacitor)
  • 2. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM Circuit Basics DRAM Cell ... Bit Lines... Memory Array RowDecoder ...WordLines... DRAM Storage element Switching element Bit Line Word Line Data In/Out Buffers Sense Amps Column Decoder (capacitor)
  • 3. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland Row, Bitlines and Wordlines DRAM Circuit Basics “Row” Defined Bit Lines Word Line “Row” of DRAM Row Size: 8 Kb @ 256 Mb SDRAM node 4 Kb @ 256 Mb RDRAM node
  • 4. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM Circuit Basics Sense Amplifier I Sense and Amplify 6 Rows shown 1 2 3 4 5 6
  • 5. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM Circuit Basics Sense Amplifier II : Precharged Vcc (logic 1) Gnd (logic 0) Vcc/2 precharged to Vcc/2 Sense and Amplify 1 2 3 4 5 6
  • 6. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM Circuit Basics Sense Amplifier III : Destructive Read Sense and Amplify 1 2 3 4 5 6 Vcc (logic 1) Gnd (logic 0) Vcc/2 Wordline Driven
  • 7. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM Access Protocol ROW ACCESS AKA: OPEN a DRAM Page/Row RAS (Row Address Strobe) or or ACT (Activate a DRAM Page/Row) BUS MEMORY CONTROLLERCPU ... Bit Lines... Memory Array RowDecoder ...WordLines... DRAM Data In/Out Buffers Sense Amps Column Decoder
  • 8. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland once the data is valid on ALL of the bit lines, you can select a subset of the bits and send them to the output buffers ... CAS picks one of the bits big point: cannot do another RAS or precharge of the lines until finished reading the column data ... can’t change the values on the bit lines or the output of the sense amps until it has been read by the memory controller DRAM Circuit Basics “Column” Defined “One Row” of DRAM Column: Smallest addressable quantity of DRAM on chip SDRAM*: column size == chip data bus width (4, 8,16, 32) RDRAM: column size != chip data bus width (128 bit fixed) 4 bit wide columns SDRAM*: get “n” columns per access. n = (1, 2, 4, 8) RDRAM: get 1 column per access. #2 #3 #4 #5#0 #1 * SDRAM means SDRAM and variants. i.e. DDR SDRAM
  • 9. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM Access Protocol COLUMN ACCESS I READ Command or CAS: Column Address Strobe BUS MEMORY CONTROLLERCPU ... Bit Lines... Memory Array RowDecoder ...WordLines... DRAM Data In/Out Buffers Sense Amps Column Decoder
  • 10. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland then the data is valid on the data bus ... depending on what you are using for in/out buffers, you might be able to overlap a litttle or a lot of the data transfer with the next CAS to the same page (this is PAGE MODE) DRAM Access Protocol Column Access II note: page mode enables overlap with CAS BUS MEMORY CONTROLLERCPU ... Bit Lines... Memory Array RowDecoder ...WordLines... DRAM Data In/Out Buffers Sense Amps Column Decoder ... with optional additional CAS: Column Address Strobe Data Out
  • 11. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland NOTE DRAM “Speed” Part I How fast can I move data from DRAM cell to sense amp? RCD (Row Command Delay) BUS MEMORY CONTROLLERCPU ... Bit Lines... Memory Array RowDecoder ...WordLines... DRAM Data In/Out Buffers Sense Amps Column Decoder tRCD
  • 12. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM “Speed” Part II How fast can I get data out of sense amps back into memory controller? BUS MEMORY CONTROLLERCPU ... Bit Lines... Memory Array RowDecoder ...WordLines... DRAM Data In/Out Buffers Sense Amps Column Decoder CAS: Column Address Strobe tCAS aka tCL tCASL aka CASL: Column Address Strobe Latency CL: Column Address Strobe Latency
  • 13. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM “Speed” Part III How fast can I move data from DRAM cell into memory controller? RAC (Random Access Delay) BUS MEMORY CONTROLLERCPU ... Bit Lines... Memory Array RowDecoder ...WordLines... DRAM Data In/Out Buffers Sense Amps Column Decoder tRAC = tRCD + tCAS
  • 14. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM “Speed” Part IV How fast can I precharge DRAM array so I can engage another RAS? RP (Row Precharge Delay) BUS MEMORY CONTROLLERCPU ... Bit Lines... Memory Array RowDecoder ...WordLines... DRAM Data In/Out Buffers Sense Amps Column Decoder tRP
  • 15. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM “Speed” Part V How fast can I read from different rows? RC (Row Cycle Time) BUS MEMORY CONTROLLERCPU ... Bit Lines... Memory Array RowDecoder ...WordLines... DRAM Data In/Out Buffers Sense Amps Column Decoder tRC = tRAS + tRP
  • 16. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM “Speed” Summary I What do I care about? RC :Row Cycle Time tRC = tRAS + tRP RP :Row Precharge Delay tRP RAC :Random Access Delay tRAC = tRCD + tCAS CAS: Column Address Strobe tCAS RAS: Row Address Strobe tRCD Seen in ads. Easy to explain Embedded systems designers DRAM manufactuers Computer Architect: Latency bound code i.e. linked list traversal Easy to sell RCD: Row Command Delay
  • 17. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM “Speed” Summary II DRAM Type Frequency Data Bus Width (per chip) Peak Data Bandwidth (per Chip) Random Access Time (tRAC) Row Cycle Time (tRC) PC133 SDRAM 133 16 200 MB/s 45 ns 60 ns DDR 266 133 * 2 16 532 MB/s 45 ns 60 ns PC800 RDRAM 400 * 2 16 1.6 GB/s 60 ns 70 ns FCRAM 200 * 2 16 0.8 GB/s 25 ns 25 ns RLDRAM 300 * 2 32 2.4 GB/s 25 ns 25 ns DRAM is “slow” But doesn’t have to be tRC < 10ns achievable Higher die cost Not commodity Expensive Not adopted in standard
  • 18. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland DRAM “latency” isn’t deterministic because of CAS or RAS+CAS, and there may be significant queuing delays within the CPU and the memory controller Each transaction has some overhead. Some types of overhead cannot be pipelined. This means that in general, longer bursts are more efficient. “DRAM latency” B C D DRAM E2/E3 E1 F A CPU Mem Controller A: Transaction request may be delayed in Queue B: Transaction request sent to Memory Controller C: Transaction converted to Command Sequences (may be queued) D: Command/s Sent to DRAM E1: Requires only a CAS or E2: Requires RAS + CAS or F: Transaction sent back to CPU “DRAM Latency” = A + B + C + D + E + F E3: Requires PRE + RAS + CAS
  • 19. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland NOTE DRAM Architecture Basics PHYSICAL ORGANIZATION This is per bank … Typical DRAMs have 2+ banks x2 DRAM x4 DRAM x8 DRAM ... Bit Lines... Sense Amps .... Data Buffers x8 DRAM RowDecoder Memory Array Column Decoder ... Bit Lines... Sense Amps ....Data Buffers x2 DRAM RowDecoder Memory Array Column Decoder ... Bit Lines... Sense Amps .... Data Buffers x4 DRAM RowDecoder Memory Array Column Decoder
  • 20. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland let’s look at the interface another way .. the say the data sheets portray it. [explain] main point: the RAS and CAS signals directly control the latches that hold the row and column addresses ... DRAM Architecture Basics Read Timing for Conventional DRAM Row Address Column Address Valid Dataout RAS CAS Address DQ Row Address Column Address Valid Dataout Data Transfer Column Access Row Access
  • 21. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland since DRAM’s inception, there have been a stream of changes to the design, from FPM to EDO to Burst EDO to SDRAM. the changes are largely structural modifications -- nimor -- that target THROUGHPUT. [discuss FPM up to SDRAM Everything up to and including SDRAM has been relatively inexpensive, especially when considering the pay-off (FPM was essentially free, EDO cost a latch, PBEDO cost a counter, SDRAM cost a slight re-design). however, we’re run out of “free” ideas, and now all changes are considered expensive ... thus there is no consensus on new directions and myriad of choices has appeared [ do LATENCY mods starting with ESDRAM ... and then the INTERFACE mods ] DRAM Evolutionary Tree (Mostly) Structural Modifications Interface Modifications Structural Conventional FPM EDO ESDRAM Rambus, DDR/2 Future Trends ........ . . . . . . MOSYS FCRAM VCDRAM $ Modifications Targeting Latency Targeting Throughput Targeting Throughput DRAM SDRAMP/BEDO
  • 22. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland NOTE DRAM Evolution Read Timing for Conventional DRAM Row Address Column Address Valid Dataout RAS CAS Address DQ Row Address Column Address Valid Dataout Data Transfer Column Access Transfer Overlap Row Access
  • 23. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland FPM aallows you to keep th esense amps actuve for multiple CAS commands ... much better throughput problem: cannot latch a new value in the column address buffer until the read-out of the data is complete DRAM Evolution Read Timing for Fast Page Mode Row Address Column Address Valid Dataout Column Address Column Address Valid Dataout Valid Dataout RAS CAS Address DQ Data Transfer Column Access Transfer Overlap Row Access
  • 24. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland solution to that problem -- instead of simple tri-state buffers, use a latch as well. by putting a latch after the column mux, the next column address command can begin sooner DRAM Evolution Read Timing for Extended Data Out Row Address Column Address Valid Dataout RAS CAS Address DQ Column Address Column Address Valid Dataout Valid Dataout Data Transfer Column Access Transfer Overlap Row Access
  • 25. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland by driving the col-addr latch from an internal counter rather than an external signal, the minimum cycle time for driving the output bus was reduced by roughly 30% DRAM Evolution Read Timing for Burst EDO Row Address Column Address RAS CAS Address DQ Data Transfer Column Access Transfer Overlap Row Access Valid Data Valid Data Valid Data Valid Data
  • 26. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland “pipeline” refers to the setting up of the read pipeline ... first CAS toggle latches the column address, all following CAS toggles drive data out onto the bus. therefore data stops coming when the memory controller stops toggling CAS DRAM Evolution Read Timing for Pipeline Burst EDO Row Address Column Address RAS CAS Address DQ Data Transfer Column Access Transfer Overlap Row Access Valid Data Valid Data Valid Data Valid Data
  • 27. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland main benefit: frees up the CPU or memory controller from having to control the DRAM’s internal latches directly ... the controller/CPU can go off and do other things during the idle cycles instead of “wait” ... even though the time-to-first-word latency actually gets worse, the scheme increases system throughput DRAM Evolution Read Timing for Synchronous DRAM (RAS + CAS + OE ... == Command Bus) Command Address DQ Clock Row Addr Col Addr Valid Data Valid Data Valid Data Valid Data ACT READ RAS CAS Data Transfer Column Access Transfer Overlap Row Access
  • 28. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland output latch on EDO allowed you to start CAS sooner for next accesss (to same row) latch whole row in ESDRAM -- allows you to start precharge & RAS sooner for thee next page access -- HIDE THE PRECHARGE OVERHEAD. DRAM Evolution Inter-Row Read Timing for ESDRAM Command Address DQ Clock Row Addr Col Addr Valid Data Valid Data Valid Data Valid Data ACT READ Row Addr Col Addr Valid Data Valid Data Valid Data Valid Data ACT READPRE “Regular” CAS-2 SDRAM, R/R to same bank Command Address DQ Clock Row Addr Col Addr Valid Data Valid Data Valid Data Valid Data ACT READ Row Addr Col Addr Valid Data Valid Data Valid Data Valid Data ACT READ ESDRAM, R/R to same bank PRE Bank Bank
  • 29. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland neat feature of this type of buffering: write-around DRAM Evolution Write-Around in ESDRAM (can second READ be this aggressive?) Command Address DQ Clock Row Addr Col Addr Valid Data Valid Data Valid Data Valid Data ACT READ Row Addr Col Addr Valid Data Valid Data Valid Data Valid Data ACT WRITEPRE “Regular” CAS-2 SDRAM, R/W/R to same bank, rows 0/1/0 Command Address DQ Clock Row Addr Col Addr Valid Data Valid Data Valid Data Valid Data ACT READ Row Addr Col Addr Valid Data Valid Data Valid Data Valid Data ACT WRITE ESDRAM, R/W/R to same bank, rows 0/1/0 PRE Bank Bank Row Addr Col Addr Valid Data Valid Data Valid Data ACT READPRE Bank Col Addr Valid Data Valid Data Valid Data Valid Data READ
  • 30. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland main thing ... it is like having a bunch of open row buffers (a la rambus), but the problem is that you must deal with the cache directly (move into and out of it), not the DRAM banks ... adds an extra couple of cycles of latency ... however, you get good bandwidth if the data you want is cache, and you can “prefetch” into cache ahead of when you want it ... originally targetted at reducing latency, now that SDRAM is CAS-2 and RCD-2, this make sense only in a throughput way DRAM Evolution Internal Structure of Virtual Channel Segment cache is software-managed, reduces energy $ Row Decoder 2Kb Segment 2Kb Segment 2Kb Segment 2Kb Segment Bank A Bank B 16 “Channels” Input/Output Buffer DQs Sel/Dec (segments) Sense Amps 2Kbit # DQs Activate Prefetch Restore Read Write
  • 31. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland FCRAM opts to break up the data array .. only activate a portion of the word line 8K rows requires 13 bits tto select ... FCRAM uses 15 (assuming the array is 8k x 1k ... the data sheet does not specify) DRAM Evolution Internal Structure of Fast Cycle RAM Reduces access time and energy/access tRCD = 15ns tRCD = 5ns 8M Array 13 bits Sense Amps 8M Array 15 bits Sense Amps SDRAM FCRAM (one clock)(two clocks) RowDecoder RowDecoder (8Kr x 1Kb) (?)
  • 32. DRAM Memory System: Lecture 2 Spring 2003 Bruce Jacob David Wang University of Maryland MoSys takes this one step further ... DRAM with an SRAM interface & speed but DRAM energy [physical partitioning: 72 banks] auto refresh -- how to do this transparently? the logic moves tthrough the arrays, refreshing them when not active. but what is one bank gets repeated access for a long duration? all other banks will be refreshed, but that one will not. solution: they have a bank-sized CACHE of lines ... in theory, should never have a problem (magic) DRAM Evolution Internal Structure of MoSys 1T-SRAM ........ . . . . . . addr Bank Select DQs $ Auto Refresh