SMR in Linux Systems Seagate's Contribution to Legacy File Systems - - PowerPoint PPT Presentation

▶
smr in linux systems
SMART_READER_LITE
LIVE PREVIEW

SMR in Linux Systems Seagate's Contribution to Legacy File Systems - - PowerPoint PPT Presentation

Jun 23, 2023 141 likes •312 views

SMR in Linux Systems Seagate's Contribution to Legacy File Systems Adrian Palmer, Drive Development Engineering SEAGATE combines DIFFERENT TECHNOLOGIES in new ways to SOLVE customer data storage CHALLENGES 2 Shingled Magnetic Recording (SMR)

slide-1
SLIDE 1

Seagate's Contribution to Legacy File Systems

SMR in Linux Systems

Adrian Palmer, Drive Development Engineering

slide-2
SLIDE 2 2

SEAGATE combines DIFFERENT TECHNOLOGIES in new ways

to SOLVE customer data storage CHALLENGES

slide-3
SLIDE 3 3 Data adapted from ASTC Technology Roadmap (http://www.idema.org/?page_id=5868)

Shingled Magnetic Recording (SMR)

Areal density growth curve – lowest cost/GB

0.1 1 10 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024

TB/in^2

BPMR (Bit Patterned Magnetic Recording) HDMR with BPMR BPMR with SMR HAMR with SMR SMR PMR PMR (Perpendicular Magnetic Recording) SMR (Shingle Magnetic Recording) HAMR (Heat Assisted Magnetic Recording)

slide-4
SLIDE 4 4 Image: Wood, R.; Williams, Mason; Kavcic, A.; Miles, Jim, "The Feasibility of Magnetic Recording at 10 Terabits Per Square Inch on Conventional Media," Magnetics, IEEE Transactions on , vol.45, no.2, pp.917,923, Feb. 2009

Shingled Magnetic Recording (SMR)

Forward-write only: Radial AND Rotational

slide-5
SLIDE 5

 Drive Managed (DM)

 Mimics Traditional drives  Backwards compatible  Direct Replacement for conventional drives in conventional apps

 Host Managed (HM)

 Not backwards compatible  Host required to manage data ordering for performance mitigation  Extensions required in ATA and SCSI command sets

 Host Aware (HA)

 Combination of DM and HM.  Backwards compatible / Able to use extensions in ATA and SCSI

SMR Drive Types

slide-6
SLIDE 6

“ “

6

Host Aware

Regular SD drive FS benefits from knowledge of media layout

Drive Managed

No change: regular SD drive

Host Managed

Requires new device & FS

Host Aware: Capacity gains like Drive Managed Performance like Conventional Drive Managed – Compatibility for Today Host Aware – Performance for Tomorrow

slide-7
SLIDE 7

 Benefits

Provides continued growth in Areal Density. Enables lower cost/GB disc drives Base of new technologies – HAMR

 Support Readiness

ZBC/ZAC specifications are nearing completion T10/T13 committees work actively progressing

 Availability

Millions of DM drives shipped! Seagate’s 8TB Archive HDD v2 drive is SMR DM in production, HA forthcoming

SMR – Can we avoid it?

ZBC: Zoned Block Commands ZAC: Zoned –device ATA Commands

slide-8
SLIDE 8 8
  • Inspired by SMR. Applicable to any media
  • Separates media into bounded zones
  • Write Pointer Zones
  • Sequential write only for Host Managed (restrictive)
  • Sequential write preferred for Host Aware (permissive)
  • New common ATA/SCSI commands
  • REPORT ZONES
  • RESET WRITE POINTER
  • OPEN/CLOSE/FINISH ZONE
  • Requires communication with FS beyond simple Read/Write

ZAC/ZBC Standards

slide-9
SLIDE 9 9
  • Optimally Directed Writes – efficient

streams

  • Excellent Reads – streaming
  • File defragmentation
  • Metadata handling
  • Backwards compatibility
  • Provide reference design for other

file systems

  • Forward-write only
  • CoW requirement
  • Zone aware (ZAC/ZBC)
  • Boundaries
  • New Commands
  • REPORT_ZONES
  • RESET_WRITE_POINTER
  • OPEN/CLOSE/FINISH ZONE
  • New algorithms
  • Defragmentation

SMR Friendly File System

Requirements Goals

slide-10
SLIDE 10 10

SMRFS -EXT4

  • Minimal

architecture changes; only rearrangement of existing data

  • Popular in

commercial storage applications

  • Build upon

strengths

  • Default FS
  • f many

distros Popular FS Stability Compatible

  • n-disk

format Market Acceptance

slide-11
SLIDE 11 12

1. cmd line arguments 2. internal handling changes 3. kernel stack changes 4. IOCTL integrations 5. Algorithm enhancements 6. Utility updates 7. Host Managed Compliance

EXT4 SMRFFS Project Scope Steps

Proposed Stages

Disk SCSI EXT4 Specify mkfs Options for HA

v1

Kernel Stack

v3

v2 v2 v2

Internal changes

v2

Kernel Integration

v4

Garbage Collection

v5

Enforce Host Managed

v7

Utilities

v6

v1 v2 v3 v4 v5 v6 v7

slide-12
SLIDE 12 13

V3 - Kernel Stack Changes, V4 Kernel Integration

Proposed Stack Changes

Disk AHCI ATA S-A-T SCSI O_Direct VFS I/O Scheduler Page Cache EXT4 VFS Examine/Enforce I/O Ordering IOCTL √ √

slide-13
SLIDE 13 14 Seagate Confidential

State of project

We’ve done …

  • Laid out design, made prototypes

Discussion at LSF…

  • Consensus of key developers

We’ve got to do …

  • A lot of work – in a short time

And we need community help!

  • Ask how to contribute and how to get sample drives at our booth!
  • Contact us after Vault at adrian.palmer@seagate.com or

timothy.r.feldman@seagate.com

slide-14
SLIDE 14 15

For Drive Managed SMR

File System Parameters Influence Performance

60 80 100 120 140 160 180 200

ext4 ext4 flex_bg ext4 dm-cow zfs btrfs ext4 ext4 flex_bg ext4 dm-cow zfs btrfs ext4 ext4 flex_bg ext4 dm-cow zfs btrfs ext4 ext4 flex_bg ext4 dm-cow zfs btrfs Mimic OS Load Mimic DVD Download 5meg files Deep Dir 1meg files Deep Dir Throughput - MB/s

Drive Managed SMR - Performance by FS Parameters, for different Workloads

driveA driveB driveC

Re-arranging file system parameters for CoW to enforce forward-write-

  • nly improves performance of a DM-SMR-enabled system
slide-15
SLIDE 15

Q&A

Thank You!

Attendees and Partners