Understanding the Organization of DRAM Subsystem Components

Slide Note
Embed
Share

Explore the intricate structure of the DRAM subsystem, including memory channels, DIMMs, ranks, chips, banks, and rows/columns. Delve into the breakdown of DIMMs, ranks, chips, and banks to comprehend the design and functioning of DRAM memory systems. Gain insights into address decoding, row/column pairs, row buffer operations, and open/closed row policies in DRAM memory controllers. Courtesy of Onur Mutlu from Carnegie Mellon University.

mallory
mallory Follow

Uploaded on May 13, 2024 | 0 Views

Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

E N D

Presentation Transcript

  1. DRAM Subsystem Organization Channel DIMM Rank Chip Bank Row/Column 1 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  2. The DRAM Subsystem Channel DIMM (Dual in-line memory module) Processor Memory channel Memory channel 2 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  3. Breaking Down a DIMM DIMM (Dual in-line memory module) Side view Front of DIMM Back of DIMM 3 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  4. Breaking Down a DIMM DIMM (Dual in-line memory module) Side view Front of DIMM Back of DIMM Rank 0: collection of 8 chips Rank 1 4 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  5. Rank Design Rank 0 (Front) Rank 1 (Back) <0:63> <0:63> Addr/Cmd CS <0:1> Data <0:63> Memory channel 5 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  6. Breaking Down a Rank . . . Chip 0 Chip 1 Chip 7 Rank 0 <56:63> <8:15> <0:7> <0:63> Data <0:63> 6 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  7. Breaking Down a Chip Chip 0 Bank 0 <0:7> <0:7> <0:7> ... <0:7> <0:7> 7 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  8. Breaking Down a Bank 2kB 1B (column) row 16k-1 ... Bank 0 row 0 <0:7> Row-buffer 1B 1B 1B ... <0:7> 8 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  9. Accessing DRAM Memory controller decodes each address into a <row,column> pair Open row policy: If the row buffer does not contain the proper row Precharge the row buffer to prepare to load the row Activate (RAS) opens row (places it into the row buffer) Read/write (CAS) reads/writes column in the row buffer Closed row policy: Activate (RAS) opens row (places it into the row buffer) Read/write (CAS) reads/writes column in the row buffer Precharge after every read/write (overlap with read/write) 9 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  10. Transferring a Cache Block Physical memory space 0xFFFF F Channel 0 ... DIMM 0 0x40 Rank 0 64B cache block 0x00 10 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  11. Transferring a Cache Block Physical memory space Chip 1 Chip 0 Chip 7 Rank 0 0xFFFF F . . . ... <56:63> <8:15> <0:7> 0x40 64B Data <0:63> cache block 0x00 11 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  12. Transferring a Cache Block Physical memory space Chip 1 Chip 0 Chip 7 Rank 0 0xFFFF F . . . Row 0 Col 0 ... <56:63> <8:15> <0:7> 0x40 64B Data <0:63> cache block 0x00 12 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  13. Transferring a cache block Physical memory space Chip 1 Chip 0 Chip 7 Rank 0 0xFFFF F . . . Row 0 Col 0 ... <56:63> <8:15> <0:7> 0x40 64B Data <0:63> cache block 8B 0x00 8B 13 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  14. Transferring a cache block Physical memory space Chip 1 Chip 0 Chip 7 Rank 0 0xFFFF F . . . Row 0 Col 1 ... <56:63> <8:15> <0:7> 0x40 64B Data <0:63> cache block 8B 0x00 14 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  15. Transferring a cache block Physical memory space Chip 1 Chip 0 Chip 7 Rank 0 0xFFFF F . . . Row 0 Col 1 ... <56:63> <8:15> <0:7> 0x40 64B Data <0:63> 8B cache block 8B 0x00 8B 15 Slide courtesy of Onur Mutlu, Carnegie Mellon University

  16. Transferring a cache block Physical memory space Chip 1 Chip 0 Chip 7 Rank 0 0xFFFF F . . . Row 0 Col 1 ... <56:63> <8:15> <0:7> 0x40 64B Data <0:63> 8B cache block 8B 0x00 A 64B cache block takes 8 I/O cycles to transfer (4 total cycles with DDR). During the process, 8 columns are read sequentially. 16 Slide courtesy of Onur Mutlu, Carnegie Mellon University

Related


Panopticon: Complete In-DRAM Rowhammer Mitigation

Panopticon: Complete In-DRAM Rowhammer Mitigation

monta
Computer Architecture: Understanding SRAM and DRAM Memory Technologies

Computer Architecture: Understanding SRAM and DRAM Memory Technologies

klay
Understanding Latency Variation in Modern DRAM Chips

Understanding Latency Variation in Modern DRAM Chips

glei
Insights into DRAM Power Consumption and Design Concerns

Insights into DRAM Power Consumption and Design Concerns

emberly
Enhancing Data Movement Efficiency in DRAM with Low-Cost Inter-Linked Subarrays (LISA)

Enhancing Data Movement Efficiency in DRAM with Low-Cost Inter-Linked Subarrays (LISA)

bradi
CLR-DRAM: Dynamic Capacity-Latency Trade-off Architecture

CLR-DRAM: Dynamic Capacity-Latency Trade-off Architecture

nurs_aws
Understanding DRAM Errors: Implications for System Design

Understanding DRAM Errors: Implications for System Design

arta766
Intelligent DRAM Cache Strategies for Bandwidth Optimization

Intelligent DRAM Cache Strategies for Bandwidth Optimization

igart
SIMDRAM: An End-to-End Framework for Bit-Serial SIMD Processing Using DRAM

SIMDRAM: An End-to-End Framework for Bit-Serial SIMD Processing Using DRAM

khadija
Enhancing Memory Cache Efficiency with DRAM Compression Techniques

Enhancing Memory Cache Efficiency with DRAM Compression Techniques

trav_45
Architecting DRAM Caches for Low Latency and High Bandwidth

Architecting DRAM Caches for Low Latency and High Bandwidth

fpiro
Enhancing DRAM Performance with ChargeCache: A Novel Approach

Enhancing DRAM Performance with ChargeCache: A Novel Approach

migisi
Quantum Error-Correcting Codes and Subsystem Codes

Quantum Error-Correcting Codes and Subsystem Codes

kard670
Understanding RowPress: A New Read Disturbance Phenomenon in Modern DRAM Chips

Understanding RowPress: A New Read Disturbance Phenomenon in Modern DRAM Chips

ami_dom
Understanding Power Consumption in Memory-Intensive Databases

Understanding Power Consumption in Memory-Intensive Databases

reeg_811
High-Throughput True Random Number Generation Using QUAC-TRNG

High-Throughput True Random Number Generation Using QUAC-TRNG

vishnu
AIV Equipment Requirements and Procedures for HERMES Payload Integration Meeting

AIV Equipment Requirements and Procedures for HERMES Payload Integration Meeting

zaydrianc
An Overview of MSPM0 Debug Subsystem

An Overview of MSPM0 Debug Subsystem

amte680
Tutorial: Installing Hadoop 3.3 on Windows 10 and Setting Up Linux Subsystem

Tutorial: Installing Hadoop 3.3 on Windows 10 and Setting Up Linux Subsystem

tassilo
Overview of Windows Subsystem for Linux 2 (WSL2)

Overview of Windows Subsystem for Linux 2 (WSL2)

jeniveve
Overview of Windows Subsystem for Linux 2 (WSL)

Overview of Windows Subsystem for Linux 2 (WSL)

gruenwald_k
Dram Shop Act and Premises Liability for Bar and Tavern Owners

Dram Shop Act and Premises Liability for Bar and Tavern Owners

juna
Improving GPGPU Performance with Cooperative Thread Array Scheduling Techniques

Improving GPGPU Performance with Cooperative Thread Array Scheduling Techniques

marmaduke
Enhancing Multi-Node Systems with Coherent DRAM Caches

Enhancing Multi-Node Systems with Coherent DRAM Caches

weld_les

More Related Content