Data Distribution Algorithms for Reliable Parallel Storage on ...
Upcoming SlideShare
Loading in...5
×
 

Data Distribution Algorithms for Reliable Parallel Storage on ...

on

  • 491 views

 

Statistics

Views

Total Views
491
Views on SlideShare
491
Embed Views
0

Actions

Likes
0
Downloads
5
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    Data Distribution Algorithms for Reliable Parallel Storage on ... Data Distribution Algorithms for Reliable Parallel Storage on ... Presentation Transcript

    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Data Distribution Algorithms for Reliable Parallel Storage on Flash Memories Kathrin Peter Zuse Institute Berlin November 2008, MEMICS Workshop Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Motivation Nonvolatile storage Flash memory - Invented by Dr. Fujio Masuoka 1984 Type of EEPROM Usage in a RAID-like configuration instead of hard disk drives Flash memory based storage: Alternative to hard disk drives? Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Outline 1 Flash based memory 2 Parallel and distributed storage based on Flash memories 3 Discussion and Summary Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Application of Flash memories Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Pros and Cons Less power consumption Higher access rates (in some cases) Price Uniform access time for random Limited erase cycles access - no seeks Flash management Robustness (extreme temperatures, vibration, shock) Model 2.5” SATA 3.0 Gbps SSD 2.5” SATA 3.0Gbps HDD Mechanism type Solid NAND flash based Magnetic rotating platters Density 64 GByte 80 GByte Weight 73 g 365g Active Power consumption 1W 3.86 W Operating temperature 0°C-70° C 5°C-55° C Acoustic Noise None 0.3 dB Endurance MTBF > 2M hours MTBF <0.7M hours Av. access time 0.1 msec 17 msec Read performance 100 MB/s 34 MB/s Write performance 80 MB/s 34 MB/s Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Limited erase cycles and flash management Transistor Control gate Insulating oxide layers Floating gate Source Drain Memory cell: floating gate n+ n+ transistor Retention and Endurance P−Substrate NAND − flash Typical page size: 16896 B page page page page block (=2 KB+64 B spare) Typical block size: block 64 pages = 128 KB =erase unit . . . Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Mapping Problem: How to map logical blocks to flash adresses? Disadvantage of linear mapping (one-to-one mapping): Frequently-used erase units wear out Identity mapping requires lots of copying to and from RAM (fixed block size). Example: flash mod. data data data data data data data data memory erase copy to RAM copy back RAM mod. data data data data data data data data modify Solution: Sophisticated block-to-flash mapping and moving around blocks: wear leveling, garbage collection Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Mapping - example Mapping of virtual block 5 to physical block 0, page 3 flash memory virtual−to−logical example: page maps page 0 1 2 3 block 7 virtual block 5 0 0 1 2 = logical block 7= 111 3 = log. erase unit 1 4 4 1 2 5 7 5 page 3 = phy. erase unit 0 page 3 2 2 logical−to−physical erase unit map 2 0 0 1 Algorithms and Data Structures for Flash Memories, Gal, Toledo, 2004 Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Wear leveling - example flash memory hot pool cold pool order in pools older ... stop aging On Efficient Wear Leveling for Large-Scale Flash-Memory Storage Systems, Li-Pin Chang, 2007 Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Garbage collection - example Free space drops below a threshold Select blocks for reclamation Copy all live pages to free pages somewhere else Change mapping entry, update to new position Erase block and allocate pages as free erase erase block x erase block x free page erase block y life page invalid page Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Facts (local) wear leveling Performance: Erase operation is slow No in-place update Controller: Efficient mapping and erase distribution Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Flash memory in a RAID-like system Aggregate higher bandwidth Reliability (redundancy for fault tolerance) Problem: uneven usage of flash memories, more writes to redundant blocks when data becomes updated Update accumulated access load Data Redundancy flash-RAIM: Redundant Array of Independent flash Memories Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Uneven distribution of writes Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary flash-RAIM and wear leveling There exist data distribution algorithms to place data evenly on a single memory Goal: long lifetime of a single memory We work on even distribution of writes across all memories in a memory array Goal: even usage of cells on all flash memories, reliable data storage, high throughput Method: global wear leveling Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Data Distribution Algorithms Staggered Striping Hierarchical dual−pool Algorithm + − + − + − Memory 1 Memory 2 Memory 3 Different starting points of algorithms Explicit placing of data Data movement after storing and local wear leveling Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Discussion of Future work Extension of the simulator to evaluate the global wear leveling algorithms Parameter study (trade-off) Define metrics to compare algorithms Lifetime of the flash-RAIM Speed Overhead Usage of traces Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Summary Use aggregated bandwidth and fault-tolerance flash-RAIM Simulator for evaluation Next steps: Implementation and evaluation of the global wear leveling algorithm Kathrin Peter Reliable Parallel Storage on Flash Memories
    • Flash based memory Parallel and distributed storage based on Flash memories Discussion and Summary Erase cycles data/ redundancy memory distribution of access frequencies 100 Disk 1 Disk 6 90 80 70 number of erase cycles 60 50 40 30 20 10 0 1.1968e+06 1.197e+06 1.1972e+061.1974e+061.1976e+061.1978e+06 1.198e+06 1.1982e+061.1984e+061.1986e+061.1988e+06 physical erase block adress Kathrin Peter Reliable Parallel Storage on Flash Memories