on the memory requirements of block interleaver for
play

On the Memory Requirements of Block Interleaver for Batched Network - PowerPoint PPT Presentation

Jan 04, 2024 •407 likes •861 views

On the Memory Requirements of Block Interleaver for Batched Network Codes Hoover H. F. Yin, Ka Hei Ng, Xishi Wang, Qi Cao, and Lucien K. L. Ng The Chinese University of Hong Kong June, 2020 Hoover H. F. Yin (CUHK) Mem. Req. of Block

  1. On the Memory Requirements of Block Interleaver for Batched Network Codes Hoover H. F. Yin, Ka Hei Ng, Xishi Wang, Qi Cao, and Lucien K. L. Ng The Chinese University of Hong Kong June, 2020 Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 1 / 13

  2. Background Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 2 / 13

  3. Background Packet loss Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 2 / 13

  4. Background Packet loss No feedback Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 2 / 13

  5. Batched Network Coding Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 3 / 13

  6. Batched Network Coding Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 3 / 13

  7. Batched Network Coding Burst loss? Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 3 / 13

  8. Batched Network Coding Burst loss? Depth of interleaver = # of batches in a block Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 3 / 13

  9. Block Interleaved Batched Network Coding batches 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 blocks 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 interleaved 1 1 1 2 2 2 3 3 3 1 1 1 2 2 2 3 3 3 blocks Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 4 / 13

  10. Block Interleaved Batched Network Coding batches 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 blocks 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 interleaved 1 1 1 2 2 2 3 3 3 1 1 1 2 2 2 3 3 3 blocks Reinterleave Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 4 / 13

  11. Block Interleaved Batched Network Coding batches 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 blocks 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 interleaved 1 1 1 2 2 2 3 3 3 1 1 1 2 2 2 3 3 3 blocks Reinterleave Temporary storage for the batches Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 4 / 13

  12. Block Interleaved Batched Network Coding batches 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 blocks 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 interleaved 1 1 1 2 2 2 3 3 3 1 1 1 2 2 2 3 3 3 blocks Reinterleave Temporary storage for the batches Lower bound on the memory requirement? Tight bound? Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 4 / 13

  13. Batch Buffer change in buffer size + L 1 + L 1 + L 1 + L 1 + L 1 incoming blocks T 1 T 1 T 1 T 1 T 1 · · · outgoing blocks T 2 T 2 · · · − L 2 − L 2 change in buffer size Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 5 / 13

  14. Batch Buffer change in buffer size + L 1 + L 1 + L 1 + L 1 + L 1 incoming blocks T 1 T 1 T 1 T 1 T 1 · · · outgoing blocks T 2 T 2 · · · − L 2 − L 2 change in buffer size Enqueue event : put L 1 batches (after deinterleaving) Dequeue event : fetch L 2 batches (for recoding and reinterleaving) Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 5 / 13

  15. Batch Buffer change in buffer size + L 1 + L 1 + L 1 + L 1 + L 1 incoming blocks T 1 T 1 T 1 T 1 T 1 · · · outgoing blocks T 2 T 2 · · · − L 2 − L 2 change in buffer size Enqueue event : put L 1 batches (after deinterleaving) Dequeue event : fetch L 2 batches (for recoding and reinterleaving) More terminologies: Total delay Delay-bounded enqueuer Valid dequeuing scheme Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 5 / 13

  16. Batch Buffer change in buffer size + L 1 + L 1 + L 1 + L 1 + L 1 incoming blocks T 1 T 1 T 1 T 1 T 1 · · · outgoing blocks T 2 T 2 · · · − L 2 − L 2 change in buffer size Enqueue event : put L 1 batches (after deinterleaving) Dequeue event : fetch L 2 batches (for recoding and reinterleaving) More terminologies: Total delay Delay-bounded enqueuer Valid dequeuing scheme Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 5 / 13

  17. Batch Buffer change in buffer size + L 1 + L 1 + L 1 + L 1 + L 1 incoming blocks T 1 T 1 T 1 T 1 T 1 · · · outgoing blocks T 2 T 2 · · · − L 2 − L 2 change in buffer size Enqueue event : put L 1 batches (after deinterleaving) Dequeue event : fetch L 2 batches (for recoding and reinterleaving) More terminologies: Total delay Delay-bounded enqueuer Valid dequeuing scheme Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 5 / 13

  18. Buffer Sizes Put or fetch first? + L 1 − L 2 Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 6 / 13

  19. Buffer Sizes Put or fetch first? Put first + L 1 − L 2 Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 6 / 13

  20. Buffer Sizes Put or fetch first? Put first Biqueue event + L 1 − L 2 Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 6 / 13

  21. Buffer Sizes Put or fetch first? Put first Biqueue event + L 1 − L 2 Before fetching: prior buffer size After fetching: buffer size Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 6 / 13

  22. Buffer Sizes Put or fetch first? Put first Biqueue event + L 1 − L 2 Before fetching: prior buffer size Memory required After fetching: buffer size Causality Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 6 / 13

  23. Bounded Buffer Size and Finite Total Delay Incoming rate > Outgoing rate ? Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 7 / 13

  24. Bounded Buffer Size and Finite Total Delay Incoming rate > Outgoing rate ? Outgoing rate > Incoming rate ? Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 7 / 13

  25. Bounded Buffer Size and Finite Total Delay Incoming rate > Outgoing rate ? Outgoing rate > Incoming rate ? L 1 / T 1 = L 2 / T 2 Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 7 / 13

  26. Bounded Buffer Size and Finite Total Delay Incoming rate > Outgoing rate ? Outgoing rate > Incoming rate ? L 1 / T 1 = L 2 / T 2 T 1 = L 1 ω and T 2 = L 2 ω Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 7 / 13

  27. Bounded Buffer Size and Finite Total Delay Incoming rate > Outgoing rate ? Outgoing rate > Incoming rate ? L 1 / T 1 = L 2 / T 2 T 1 = L 1 ω and T 2 = L 2 ω d timeslots = total delay d ′ timeslots = total incoming delay Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 7 / 13

  28. Bounded Buffer Size and Finite Total Delay Incoming rate > Outgoing rate ? Outgoing rate > Incoming rate ? L 1 / T 1 = L 2 / T 2 T 1 = L 1 ω and T 2 = L 2 ω d timeslots = total delay d ′ timeslots = total incoming delay Theorem The minimum total delay among all valid dequeuing schemes is L 1 + L 2 − gcd( L 1 , L 2 ) + d ′ timeslots. Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 7 / 13

  29. Periodic Structure: L 1 = 8 , L 2 = 6 in s1 s2 s1 prior buffer size 16 s2 12 10 t 0 52 76 100 124 148 Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 8 / 13

  30. Periodic Structure: L 1 = 8 , L 2 = 6 in s1 s2 s1 prior buffer size 16 s2 12 10 t 0 52 76 100 124 148 Exists scheme with periods Exists scheme achieving min total delay and lowest memory requirement Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 8 / 13

  31. � Periodic Structure: L 1 = 8 , L 2 = 6 in s1 s2 s1 prior buffer size 16 s2 12 10 t 0 52 76 100 124 148 Exists scheme with periods Exists scheme achieving min total delay and lowest memory requirement Same total delay = ⇒ same memory requirement Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 8 / 13

  32. Periodic Structure: L 1 = 8 , L 2 = 6 in s3 s4 s3 prior buffer size 16 s4 12 t 0 52 76 100 124 148 Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 9 / 13

  33. � Periodic Structure: L 1 = 8 , L 2 = 6 in s3 s4 s3 prior buffer size 16 s4 12 t 0 52 76 100 124 148 Exists scheme with no period and no biqueue event in a long run Lowest memory requirement = ⇒ min total delay Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 9 / 13

  34. � � Periodic Structure: L 1 = 8 , L 2 = 6 in s3 s4 s3 prior buffer size 16 s4 12 t 0 52 76 100 124 148 Exists scheme with no period and no biqueue event in a long run Lowest memory requirement = ⇒ min total delay Exists period = ⇒ exists biqueue event in the period Peak is not always unique Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 9 / 13

  35. Periodic Structure: L 1 = 8 , L 2 = 6 in s5 s6 s5 prior buffer size 16 s6 14 12 2 t 0 52 76 100 124 148 Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 10 / 13

  36. � Periodic Structure: L 1 = 8 , L 2 = 6 in s5 s6 s5 prior buffer size 16 s6 14 12 2 t 0 52 76 100 124 148 Same total delay = ⇒ same lower bound on the memory requirement Hoover H. F. Yin (CUHK) Mem. Req. of Block Interleaver for BNC June, 2020 10 / 13

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

General Cache Mechanics CPU Block: unit of data in cache and memory. (a.k.a. line) Memory

General Cache Mechanics CPU Block: unit of data in cache and memory. (a.k.a. line) Memory

General Cache Mechanics CPU Block: unit of data in cache and memory. (a.k.a. line) Memory Hierarchy: Cache Smaller, faster, more expensive. Cache 8 9 14 3 Stores subset of memory blocks . (lines) Data is moved in block units Memory

467 views • 8 slides
Memory Management Memory Manager Requirements Minimize primary memory access time

Memory Management Memory Manager Requirements Minimize primary memory access time

Memory Management Memory Manager Requirements Minimize primary memory access time Maximize primary memory size Primary memory must be cost-effective Todays memory manager: Allocates primary memory to processes Maps

637 views • 27 slides
CSE 351: Section 10 Memory Allocation Memory Allocation Must allocate any memory you need to

CSE 351: Section 10 Memory Allocation Memory Allocation Must allocate any memory you need to

CSE 351: Section 10 Memory Allocation Memory Allocation Must allocate any memory you need to use Allocating memory = designating a block of it as 'used' This way, you don't overwrite your data The memory isn't changed, but

417 views • 15 slides
1 3. Simple Memory System Cache Address Translation Example #1 16 lines, 4-byte block size

1 3. Simple Memory System Cache Address Translation Example #1 16 lines, 4-byte block size

Today Simple memory system example Case study: Core i7/Linux memory system Virtual Memory: Systems Memory mapping CSci 2021: Machine Architecture and Organization April 20th-22nd, 2020 Your instructor: Stephen McCamant Based on

587 views • 5 slides
Section 11 Section 11 Direct Memory Access (DMA) a 11-1 1 ADSP-BF533 Block Diagram L1

Section 11 Section 11 Direct Memory Access (DMA) a 11-1 1 ADSP-BF533 Block Diagram L1

Section 11 Section 11 Direct Memory Access (DMA) a 11-1 1 ADSP-BF533 Block Diagram L1 Core Instruction Timer 64 Memory Performance Core LD0 32 Monitor Processor L1 Data LD1 32 Memory JTAG/ Debug SD32 Core D0 bus Core I bus

668 views • 37 slides
CUDA Programming Model Ming Yang Apr 5, 2016 Thread Local memory Warp Occupancy Block Grid

CUDA Programming Model Ming Yang Apr 5, 2016 Thread Local memory Warp Occupancy Block Grid

CUDA Programming Model Ming Yang Apr 5, 2016 Thread Local memory Warp Occupancy Block Grid Shared memory Kernel Register Global memory 1 ??? What are the scheduling units in Streaming Multiprocessor (SM)?? warps. How are they

513 views • 28 slides
Block Device Scheduling Don Porter CSE 506 Logical Diagram Binary Memory Threads Formats

Block Device Scheduling Don Porter CSE 506 Logical Diagram Binary Memory Threads Formats

Block Device Scheduling Don Porter CSE 506 Logical Diagram Binary Memory Threads Formats Allocators User System Calls Kernel RCU File System Networking Sync Memory CPU Device Management Scheduler Drivers Todays Lecture

694 views • 36 slides
Today Memory Management Virtual Memory Motivation and Requirements Project 5 Nov 28,

Today Memory Management Virtual Memory Motivation and Requirements Project 5 Nov 28,

Today Memory Management Virtual Memory Motivation and Requirements Project 5 Nov 28, 2018 Sprenkle - CSCI330 1 Review What is RAID? What is its motivation? What are its goals? What are some RAID levels? What

580 views • 19 slides
Main Memory Moving further away from the CPU .. 95 Main Memory Performance measurement

Main Memory Moving further away from the CPU .. 95 Main Memory Performance measurement

Main Memory Moving further away from the CPU .. 95 Main Memory Performance measurement Latency - cache miss penalty Bandwidth - large block sizes of L2 argue for B/W Memory latency Access time : Time between when a read is

652 views • 14 slides
Evaluating Grant Applications with Generalized Chain Block Designs in R Dedicated to the Memory

Evaluating Grant Applications with Generalized Chain Block Designs in R Dedicated to the Memory

Evaluating Grant Applications with Generalized Chain Block Designs in R Dedicated to the Memory of John Mandel Giles Crane, Cynthia Collins, and Karin Mille NJ Department of Health and Senior Services 1 Grant Application Review via

572 views • 33 slides
Today Memory Management Segmentation, Paging Improving memory performance MMU

Today Memory Management Segmentation, Paging Improving memory performance MMU

Today Memory Management Segmentation, Paging Improving memory performance MMU Translation Lookaside Buffer Dec 3, 2018 Sprenkle - CSCI330 1 Review What abstraction does virtual memory provide? What requirements do we

661 views • 25 slides
Main Memory Tevfik Ko ar Louisiana State University February 28 th , 2008 1 Memory

Main Memory Tevfik Ko ar Louisiana State University February 28 th , 2008 1 Memory

CSC 4103 - Operating Systems Spring 2008 Lecture - XI Main Memory Tevfik Ko ar Louisiana State University February 28 th , 2008 1 Memory Management Requirements The O/S must fit multiple processes in memory memory needs to be

467 views • 11 slides
EE 457 Unit 7b Main Memory Organization 2 Motivation Organize main memory to Facilitate

EE 457 Unit 7b Main Memory Organization 2 Motivation Organize main memory to Facilitate

1 EE 457 Unit 7b Main Memory Organization 2 Motivation Organize main memory to Facilitate byte-addressability while maintaining Efficient fetching of the words in a cache block Low order interleaving (L.O.I) helps us achieve

799 views • 31 slides
1 Implementation Snoop Caches Implementing Snooping Caches Write Races: Multiple processors

1 Implementation Snoop Caches Implementing Snooping Caches Write Races: Multiple processors

An Example Snoopy Protocol Invalidation protocol, write-back cache Each block of memory is in one state: Cache Coherence and Memory Clean in all caches and up-to-date in memory (Shared) Consistency OR Dirty in exactly one cache

507 views • 3 slides
Memory II. Memory improvement III. Problems with memory 3 systems/stages of Memory: memory

Memory II. Memory improvement III. Problems with memory 3 systems/stages of Memory: memory

Main Focus I. Memory as a process Memory II. Memory improvement III. Problems with memory 3 systems/stages of Memory: memory the process by which I. Sensory Memory information is - acquired, II. Short -Term Memory - stored,

169 views • 5 slides
Cache Systems CPU Main Main CPU Memory Memory 400MHz 10MHz Cache 10MHz Memory Hierarchy

Cache Systems CPU Main Main CPU Memory Memory 400MHz 10MHz Cache 10MHz Memory Hierarchy

Cache Systems CPU Main Main CPU Memory Memory 400MHz 10MHz Cache 10MHz Memory Hierarchy Design Bus 66MHz Bus 66MHz Chapter 5 and Appendix C Data object Block transfer transfer Main CPU Cache Memory 1 4 Example: Two-level

353 views • 10 slides
Problem 1 Write the best title lines for the functions that are called by the following main

Problem 1 Write the best title lines for the functions that are called by the following main

Problem 1 Write the best title lines for the functions that are called by the following main program. Do not supply the blocks for the functions. int main() { int x = 0, y = 1, z = 2; double b[3] = {1.9, 2.3, 3.0}; int d[2][2] = {{1,2},{3,4}};

1.65k views • 126 slides
Semi-Automatic Floating-Point Implementation of Special Functions Christoph Lauter 1 Marc

Semi-Automatic Floating-Point Implementation of Special Functions Christoph Lauter 1 Marc

Semi-Automatic Floating-Point Implementation of Special Functions Christoph Lauter 1 Marc Mezzarobba 1,2 Pequan group 1 Universit Paris 6 2 CNRS ARITH 22, Lyon, 2015-06-23 Code Generation for Mathematical Functions expression explicit

465 views • 25 slides
CSCI 144 - Introduction to Computer Science Instructor: John Goettsche Computer Science

CSCI 144 - Introduction to Computer Science Instructor: John Goettsche Computer Science

CSCI 144 - Introduction to Computer Science Instructor: John Goettsche Computer Science Department Pacific Lutheran University Spring 20018 1 Reading questions Simple website: http://cs.plu.edu/~caora/ Enter your session number, and

419 views • 41 slides
The network monitoring in grid context Operations Perspective Emir Imamagic /SRCE EGEE09,

The network monitoring in grid context Operations Perspective Emir Imamagic /SRCE EGEE09,

Enabling Grids for E-sciencE The network monitoring in grid context Operations Perspective Emir Imamagic /SRCE EGEE09, Barcelona, Spain www.eu-egee.org

440 views • 17 slides
Optimizing Motion-to-Photon Latency on DAQRI Smart Glasses Heinrich Fink

Optimizing Motion-to-Photon Latency on DAQRI Smart Glasses Heinrich Fink

Optimizing Motion-to-Photon Latency on DAQRI Smart Glasses Heinrich Fink heinrich.fink@daqri.com XDC 2018 1 System Overview Himax Intel SKL Gen9 Display-Port Ctrl + LCOS m7-6Y75 Hardware Platform Software Linux 4.14.x Motion Samples

105 views • 7 slides
Paper presentation Ultra-Portable Devices Paper: S. Hong, W. E. Stark. Decoding Performance

Paper presentation Ultra-Portable Devices Paper: S. Hong, W. E. Stark. Decoding Performance

Paper presentation Ultra-Portable Devices Paper: S. Hong, W. E. Stark. Decoding Performance and Complexity Analysis for Analog and Digital Channel Decoders. Vehicular Technology Conference (VTC), May 2001, vol. 2, pages 1277-1281.

149 views • 14 slides
Delayed cord clamping: Studied with Judith Mercer Not just for preemies *Drop the

Delayed cord clamping: Studied with Judith Mercer Not just for preemies *Drop the

Nothing to disclose (kind of) No financial ties Delayed cord clamping: Studied with Judith Mercer Not just for preemies *Drop the scissors* Kate Norman Fr meta, MS, CNM Assistant Professor, UCSF Division of Maternal-Fetal

555 views • 9 slides
Incentive-Compatible Differentiated Scheduling Background 2 HotNets IV - November 2005

Incentive-Compatible Differentiated Scheduling Background 2 HotNets IV - November 2005

Incentive-Compatible Differentiated Scheduling Background 2 HotNets IV - November 2005 Scheduler Model 4 Implementation 6 Evaluation 8 Discussion 12 Wrap Up 13 Martin Karsten, Yunfeng Lin, Kate Larson School of Computer Science,

339 views • 14 slides

More recommend