pith. sign in

arxiv: 1907.11825 · v1 · pith:QAKKIEEHnew · submitted 2019-07-27 · 💻 cs.OS

SSDFS: Towards LFS Flash-Friendly File System without GC operation

Viacheslav Dubeyko This is my paper

Pith reviewed 2026-05-24 15:18 UTC · model grok-4.3

classification 💻 cs.OS
keywords SSDFSfile systemSSDflash storagewrite amplificationgarbage collectionlog-structured file systemDiff-On-Write
0
0 comments X

The pith

SSDFS combines logical segments, diff-on-write, and specialized b-trees to manage SSD writes without traditional garbage collection.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents SSDFS as a log-structured file system built for flash storage. It defines a collection of new structures such as logical segments that group erase blocks, a diff-on-write update method, PEB migration, and multiple b-tree types for metadata and extents. The central argument is that these pieces together control how data is placed and updated so that write amplification stays low, garbage collection work shrinks, device lifetime lengthens, and I/O performance remains steady. A reader would care because existing flash file systems often force extra NAND overprovisioning and produce unpredictable slowdowns as the drive ages. If the mechanisms integrate cleanly, the design points toward simpler, longer-lasting SSD usage in everyday systems.

Core claim

SSDFS file system introduces several authentic concepts and mechanisms: logical segment, logical extent, segment's PEBs pool, Main/Diff/Journal areas in the PEB's log, Diff-On-Write approach, PEBs migration scheme, hot/warm data self-migration, segment bitmap, hybrid b-tree, shared dictionary b-tree, shared extents b-tree. Combination of all suggested concepts are able: (1) manage write amplification in smart way, (2) decrease GC overhead, (3) prolong SSD lifetime, and (4) provide predictable file system's performance.

What carries the argument

The Diff-On-Write approach inside Main/Diff/Journal areas of PEB logs, paired with logical segments, PEB migration, and hybrid/shared b-trees that track extents and dictionaries.

If this is right

  • Write amplification is kept low by writing only changed portions of data rather than full blocks.
  • GC overhead drops because hot and warm data self-migrate within logical segments.
  • SSD lifetime extends because fewer total writes reach the NAND cells.
  • File system performance stays predictable because background migration replaces sudden GC bursts.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The segment bitmap and PEB pool could be reused in other log-structured systems to simplify extent tracking.
  • Shared dictionary and extents b-trees might reduce metadata duplication in large-scale storage setups.
  • The migration scheme for hot data could be tested as an add-on to existing flash file systems to measure lifetime gains.

Load-bearing premise

The listed mechanisms can be combined in an actual implementation to deliver the four listed benefits without introducing offsetting overheads or compatibility problems.

What would settle it

A working prototype of SSDFS run on real SSD hardware that exhibits higher write amplification, more frequent GC pauses, or shorter measured endurance than F2FS under identical mixed read-write workloads would falsify the central claim.

Figures

Figures reproduced from arXiv: 1907.11825 by Viacheslav Dubeyko.

Figure 1
Figure 1. Figure 1: Logical segment concept. Logical segment. Generally speaking, segment would represent the real physical unit(s) (for example, one or sev￾eral PEBs are identified by LBAs on the storage device). However, SSDFS operates by logical segments. The logi￾cal segment is the unit that is always located on some offset from the volume’s beginning for the whole lifetime of file system volume ( [PITH_FULL_IMAGE:figure… view at source ↗
Figure 2
Figure 2. Figure 2: Logical extent concept. Logical extent. Usually, segment is associated with a PEB (flash-oriented file system) or with a LBA (flash￾friendly file system). However, segment is the pure logical entity without the strict relation with PEB or LBA in the case of SSDFS file system. Generally speaking, the segment is simply some portion of the file system volume is always lo￾cated on some offset from the volumes … view at source ↗
Figure 3
Figure 3. Figure 3: Segment parallelism. Segment parallelism. One of the important goal to have several LEBs/PEBs in one segment is the trying to employ the parallelism of operation with PEBs are located on dif￾ferent dies. Usually, any SSD contains a set of dies are able to execute various operations independently and con￾currently (for example, erase operation). Moreover, multi￾channel SSD architecture is capable to deliver… view at source ↗
Figure 4
Figure 4. Figure 4: Log concept. It is possible to imagine the log like a container that in￾cludes a header, a payload, and a footer ( [PITH_FULL_IMAGE:figures/full_fig_p019_4.png] view at source ↗
Figure 7
Figure 7. Figure 7: Log footer. Superblock. Usually, any file system starts from a su￾perblock that is located in one or several fixed position(s) on the file system’s volume. The responsibility of the su￾perblock is to identify the file system’s type and to provide the description of the key file system’s metadata structures. SSDFS represents the LFS file system type that is using the Copy-On-Write (COW) policy for updating … view at source ↗
Figure 5
Figure 5. Figure 5: Log header. If it exists the necessity to commit a log without the pres￾ence of enough data in the payload then it needs to create a chain of partial logs in a PEB ( [PITH_FULL_IMAGE:figures/full_fig_p019_5.png] view at source ↗
Figure 9
Figure 9. Figure 9: Technique of using the block bitmap. Block bitmap is the PEB-based metadata structure in the case of SSDFS file system ( [PITH_FULL_IMAGE:figures/full_fig_p020_9.png] view at source ↗
Figure 8
Figure 8. Figure 8: Block bitmap concept. SSDFS file system introduces the PEB-based block bitmap because of proven efficiency and compactness of this metadata structure. First of all, the block bitmap ( [PITH_FULL_IMAGE:figures/full_fig_p020_8.png] view at source ↗
Figure 10
Figure 10. Figure 10: Offsets translation table concept. Offsets translation table. Any subsystem of SSDFS file system’s driver that needs to store user data or metadata 20 [PITH_FULL_IMAGE:figures/full_fig_p020_10.png] view at source ↗
Figure 12
Figure 12. Figure 12: Log structure. Log structure. As a result, log’s structure ( [PITH_FULL_IMAGE:figures/full_fig_p021_12.png] view at source ↗
Figure 11
Figure 11. Figure 11: Offsets translation table architecture. Generally speaking, the offsets translation table includes several metadata structures inside of the log ( [PITH_FULL_IMAGE:figures/full_fig_p021_11.png] view at source ↗
Figure 13
Figure 13. Figure 13: Main, diff and journal payload areas. the journal area. The goal of journal area ( [PITH_FULL_IMAGE:figures/full_fig_p022_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Classic superblock approach [PITH_FULL_IMAGE:figures/full_fig_p022_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Distributed superblock approach. Any SSDFS file system’s volume represents a sequence of logical segments. Every segment contains some number of LEBs. Finally, it needs to associate a LEB with a PEB in the case of necessity to store any data in the segment. As a result, 22 [PITH_FULL_IMAGE:figures/full_fig_p022_15.png] view at source ↗
Figure 17
Figure 17. Figure 17: Superblock segments’ migration scheme. actual superblock segments or more actual numbers of su￾perblock segments. Finally, it is possible to find the actual superblock segment by means of passing through the chain of segment numbers. As a result, it needs to find the lat￾est log in the found actual superblock segment with the goal to retrieve the actual superblock’s state. Moreover, SSDFS file system keep… view at source ↗
Figure 16
Figure 16. Figure 16: Specialized superblock concept. However, the massive replication of superblock’s meta￾data creates the problem to find the last actual state of mu￾table part of superblock’s metadata. To resolve this problem the SSDFS file system introduces a special type of segment - the superblock segment ( [PITH_FULL_IMAGE:figures/full_fig_p023_16.png] view at source ↗
Figure 20
Figure 20. Figure 20: Snapshots concept. Finally, snapshot table in the superblock segment is ca￾pable to associate the snapshot IDs with segment numbers ( [PITH_FULL_IMAGE:figures/full_fig_p024_20.png] view at source ↗
Figure 18
Figure 18. Figure 18: Snapshots table concept [PITH_FULL_IMAGE:figures/full_fig_p024_18.png] view at source ↗
Figure 19
Figure 19. Figure 19: Snapshot segment concept. SSDFS file system introduces the concept of specialized snapshot segment ( [PITH_FULL_IMAGE:figures/full_fig_p024_19.png] view at source ↗
Figure 21
Figure 21. Figure 21: PEB mapping table architecture. ping table. Because SSDFS file system employs the concept of logical segment then the reserved numbers of specialized segments remain the same for the volume’s lifetime. But if some PEB achieves the exhausted state then it triggers the migration mechanism of moving the exhausted PEB into an￾other one. Also PEB mapping table is enhanced by special cache is stored in the payl… view at source ↗
Figure 23
Figure 23. Figure 23: Possible PEB’s types and states. PEB type ( [PITH_FULL_IMAGE:figures/full_fig_p025_23.png] view at source ↗
Figure 22
Figure 22. Figure 22: PEB mapping table’s fragment structure. Every fragment of PEB mapping table represents the log’s payload in a specialized segment ( [PITH_FULL_IMAGE:figures/full_fig_p025_22.png] view at source ↗
Figure 24
Figure 24. Figure 24: PEB mapping table’s cache. PEB state, and (3) PEB flags. Generally speaking, the con￾sistency field simply shows that a record in the cache and mapping table is identical or not. If some record in the cache has marked as inconsistent then it means that the PEB map￾ping table has to be modified with the goal to keep the actual value of the cache. As a result, finally, the value in the table and the cache w… view at source ↗
Figure 26
Figure 26. Figure 26: Segment bitmap architecture. Segment bitmap is implemented like the bitmap metadata structure that is split on several fragments ( [PITH_FULL_IMAGE:figures/full_fig_p026_26.png] view at source ↗
Figure 28
Figure 28. Figure 28: B-tree segment type [PITH_FULL_IMAGE:figures/full_fig_p027_28.png] view at source ↗
Figure 29
Figure 29. Figure 29: User data segment type. User data segment. SSDFS file system aggregates user data inside of segments are dedicated to user data’s type ( [PITH_FULL_IMAGE:figures/full_fig_p027_29.png] view at source ↗
Figure 30
Figure 30. Figure 30: Current segment concept [PITH_FULL_IMAGE:figures/full_fig_p028_30.png] view at source ↗
Figure 31
Figure 31. Figure 31: Common b-tree architecture. Why b-tree for LFS file system? Usually, b-tree is considered like not very good choice for the case of flash￾oriented and flash-friendly file systems by virtue of wander￾ing tree issue and high value of write amplification. How￾ever, b-tree architecture implements very important advan￾tages: (1) efficient search mechanism, (2) compact storage of sparse data, (3) flexible techn… view at source ↗
Figure 33
Figure 33. Figure 33: Hybrid node architecture [PITH_FULL_IMAGE:figures/full_fig_p030_33.png] view at source ↗
Figure 34
Figure 34. Figure 34: Hybrid b-tree evolution. The exhaustion of the data area’s space of the first hybrid node triggers addition of the second hybrid node on the first level of the b-tree ( [PITH_FULL_IMAGE:figures/full_fig_p030_34.png] view at source ↗
Figure 32
Figure 32. Figure 32: B-tree architecture with hybrid nodes. SSDFS file system uses a hybrid b-tree architecture ( [PITH_FULL_IMAGE:figures/full_fig_p030_32.png] view at source ↗
Figure 35
Figure 35. Figure 35: Hybrid b-tree evolution. of temporary buffer that aggregates enough data records be￾fore a leaf node creation. Generally speaking, this sequence of leaf nodes creation takes place before the exhaustion of index area of hybrid node. Moreover, the index area’s ex￾haustion triggers the increasing of index area’s capacity. As a result, it means decreasing the capacity of data area in hy￾brid node. If the inde… view at source ↗
Figure 36
Figure 36. Figure 36: Node type migration scheme. Operation of deletion of data records could initiate the transformation of index node(s) into the hybrid ones ( [PITH_FULL_IMAGE:figures/full_fig_p031_36.png] view at source ↗
Figure 37
Figure 37. Figure 37: Inodes b-tree architecture. Generally speaking, inode table can be imagined like a generalized array of inode instances ( [PITH_FULL_IMAGE:figures/full_fig_p032_37.png] view at source ↗
Figure 38
Figure 38. Figure 38: Raw inode structure. SSDFS raw inode ( [PITH_FULL_IMAGE:figures/full_fig_p032_38.png] view at source ↗
Figure 39
Figure 39. Figure 39: Dentries b-tree architecture. SSDFS dentry ( [PITH_FULL_IMAGE:figures/full_fig_p033_39.png] view at source ↗
Figure 40
Figure 40. Figure 40: Extents b-tree architecture. SSDFS raw extent ( [PITH_FULL_IMAGE:figures/full_fig_p033_40.png] view at source ↗
Figure 41
Figure 41. Figure 41: Deduplication mechanism of shared extents b [PITH_FULL_IMAGE:figures/full_fig_p034_41.png] view at source ↗
Figure 42
Figure 42. Figure 42: Record types in shared extents b-tree. SSDFS shared extents b-tree is able to store several record types ( [PITH_FULL_IMAGE:figures/full_fig_p034_42.png] view at source ↗
Figure 43
Figure 43. Figure 43: Shared extents b-tree architecture. it contains fingerprint, extent (segment ID, logical block, length), and reference counter values. Generally speaking, the goal of these records is to find the deduplicated extents on the basis of fingerprint value. The fingerprint records are ordered by segment ID and logical block values and the responsibility of such records is to provide the way to find the fingerpr… view at source ↗
Figure 44
Figure 44. Figure 44: Shared dictionary b-tree architecture. Shared dictionary is the hybrid b-tree with root node is stored into the superblock ( [PITH_FULL_IMAGE:figures/full_fig_p035_44.png] view at source ↗
Figure 45
Figure 45. Figure 45: Names deduplication mechanism [PITH_FULL_IMAGE:figures/full_fig_p036_45.png] view at source ↗
Figure 46
Figure 46. Figure 46: Deduplicated strings representation. (2) lookup table2, (3) hash table, and (4) strings area ( [PITH_FULL_IMAGE:figures/full_fig_p036_46.png] view at source ↗
Figure 48
Figure 48. Figure 48: Extended attributes (xattr) b-tree architecture. [PITH_FULL_IMAGE:figures/full_fig_p036_48.png] view at source ↗
Figure 49
Figure 49. Figure 49: Extended attributes b-tree’s node structure. [PITH_FULL_IMAGE:figures/full_fig_p037_49.png] view at source ↗
Figure 50
Figure 50. Figure 50: Copy-On-Write policy side effect [PITH_FULL_IMAGE:figures/full_fig_p038_50.png] view at source ↗
Figure 51
Figure 51. Figure 51: Write amplification issue. Diff-On-Write approach ( [PITH_FULL_IMAGE:figures/full_fig_p038_51.png] view at source ↗
Figure 52
Figure 52. Figure 52: Diff-On-Write approach [PITH_FULL_IMAGE:figures/full_fig_p039_52.png] view at source ↗
Figure 53
Figure 53. Figure 53: Copy-On-Write vs. Diff-On-Write [PITH_FULL_IMAGE:figures/full_fig_p039_53.png] view at source ↗
Figure 56
Figure 56. Figure 56: Technique of main and journal areas interaction [PITH_FULL_IMAGE:figures/full_fig_p040_56.png] view at source ↗
Figure 57
Figure 57. Figure 57: Technique of journal and diff updates areas inter [PITH_FULL_IMAGE:figures/full_fig_p040_57.png] view at source ↗
read the original abstract

Solid state drives have a number of interesting characteristics. However, there are numerous file system and storage design issues for SSDs that impact the performance and device endurance. Many flash-oriented and flash-friendly file systems introduce significant write amplification issue and GC overhead that results in shorter SSD lifetime and necessity to use the NAND flash overprovisioning. SSDFS file system introduces several authentic concepts and mechanisms: logical segment, logical extent, segment's PEBs pool, Main/Diff/Journal areas in the PEB's log, Diff-On-Write approach, PEBs migration scheme, hot/warm data self-migration, segment bitmap, hybrid b-tree, shared dictionary b-tree, shared extents b-tree. Combination of all suggested concepts are able: (1) manage write amplification in smart way, (2) decrease GC overhead, (3) prolong SSD lifetime, and (4) provide predictable file system's performance.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript proposes SSDFS, a log-structured file system for SSDs, introducing mechanisms including logical segments, logical extents, segment PEBs pools, Main/Diff/Journal areas, Diff-On-Write, PEB migration, hot/warm self-migration, segment bitmaps, hybrid B-trees, shared dictionary B-trees, and shared extents B-trees. It claims that the combination of these concepts manages write amplification, decreases GC overhead, prolongs SSD lifetime, and yields predictable performance without GC operations.

Significance. If the mechanisms can be shown to interact without offsetting costs, the design would address longstanding write-amplification and endurance problems in flash file systems by eliminating GC, potentially improving both device lifetime and I/O predictability over conventional LFS and F2FS-style approaches.

major comments (3)
  1. [Abstract] Abstract: the central claim that the listed mechanisms 'are able' to deliver the four benefits (write-amplification management, GC elimination, lifetime extension, predictable performance) is presented without any analytical model, cost accounting, or interaction analysis showing that bitmap maintenance, multi-tree lookups, and migration traffic do not re-introduce write amplification or latency variance.
  2. [Abstract] The manuscript describes each mechanism in isolation but supplies no quantitative evaluation, simulation, or prototype measurements that would validate the claim that their combination avoids offsetting overheads (reader’s weakest assumption).
  3. [Abstract] No section provides even a high-level accounting of how Diff-On-Write plus PEB migration plus segment bitmap together eliminate GC while preserving the log-structured property; the absence of such reasoning makes the 'without GC operation' title claim unsupported.
minor comments (2)
  1. [Abstract] Abstract contains a subject-verb agreement error: 'Combination of all suggested concepts are able' should be 'is able'.
  2. The manuscript would benefit from explicit comparison tables or diagrams contrasting SSDFS mechanisms against existing LFS designs (e.g., F2FS, NILFS) on write-amplification and GC metrics.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive feedback. We respond point-by-point to the major comments and indicate planned revisions to strengthen the presentation of the design claims.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that the listed mechanisms 'are able' to deliver the four benefits (write-amplification management, GC elimination, lifetime extension, predictable performance) is presented without any analytical model, cost accounting, or interaction analysis showing that bitmap maintenance, multi-tree lookups, and migration traffic do not re-introduce write amplification or latency variance.

    Authors: We agree the abstract would benefit from explicit reference to supporting analysis. The manuscript details mechanism interactions in the design sections, but we will revise the abstract to note the cost accounting and add a dedicated subsection on interaction analysis to address potential overheads from bitmaps, lookups, and migrations. revision: yes

  2. Referee: [Abstract] The manuscript describes each mechanism in isolation but supplies no quantitative evaluation, simulation, or prototype measurements that would validate the claim that their combination avoids offsetting overheads (reader’s weakest assumption).

    Authors: The manuscript is design-focused. We acknowledge the value of validation and will add high-level analytical models plus preliminary simulation results in the revision to demonstrate that the combined mechanisms avoid offsetting overheads. revision: yes

  3. Referee: [Abstract] No section provides even a high-level accounting of how Diff-On-Write plus PEB migration plus segment bitmap together eliminate GC while preserving the log-structured property; the absence of such reasoning makes the 'without GC operation' title claim unsupported.

    Authors: We will insert a new subsection providing the requested high-level accounting. It will step through the combined operation of Diff-On-Write, PEB migration, and segment bitmaps to show GC elimination while retaining the log-structured property, thereby supporting the title claim. revision: yes

Circularity Check

0 steps flagged

No significant circularity: high-level design proposal without derivations or equations

full rationale

The manuscript is a high-level design proposal that enumerates mechanisms (logical segment, Diff-On-Write, hybrid b-tree, etc.) and asserts their combination can achieve four benefits. It contains no equations, fitted parameters, mathematical derivations, or load-bearing self-citations. No step reduces a claimed result to its own inputs by construction; the central assertion remains an untested design hypothesis rather than a circular derivation. This matches the default expectation of no circularity for papers lacking quantitative chains.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 3 invented entities

The paper is a design proposal whose central claim rests on the domain assumption that flash memory behavior can be improved by the listed new structures. No numeric free parameters are described. The new design elements function as invented entities whose independent evidence is absent from the abstract.

axioms (1)
  • domain assumption Flash memory exhibits write amplification and limited endurance that file-system organization can materially reduce.
    Implicit foundation for all claimed benefits.
invented entities (3)
  • logical segment no independent evidence
    purpose: Organize data to avoid traditional GC
    New structure introduced in the abstract.
  • Diff-On-Write approach no independent evidence
    purpose: Record changes as differences rather than full rewrites
    New write method proposed in the abstract.
  • PEBs migration scheme no independent evidence
    purpose: Move data between physical erase blocks based on temperature
    New migration mechanism proposed in the abstract.

pith-pipeline@v0.9.0 · 5678 in / 1319 out tokens · 29432 ms · 2026-05-24T15:18:16.302420+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

140 extracted references · 140 canonical work pages

  1. [1]

    Available: http://www

    SSDFS Project, [Online]. Available: http://www. ssdfs.org, Accessed on: Jun. 19, 2019

    work page 2019

  2. [2]

    V . A. Dubeyko, C. Guyot, ”Systems and methods for improving flash-oriented file system garbage collec- tion,” U.S. Patent Application US20170017405, pub- lished January 19, 2017

    work page 2017

  3. [3]

    V . A. Dubeyko, C. Guyot, ”Systems and methods for improving flash-oriented file system garbage collec- tion,” U.S. Patent Application US20170017406, pub- lished January 19, 2017

    work page 2017

  4. [4]

    V . A. Dubeyko, C. Guyot, ”Method of decreasing write amplification factor and over-provisioning of NAND flash by means of Diff-On-Write approach,” U.S. Patent Application US20170139616, published May 18, 2017

    work page 2017

  5. [5]

    V . A. Dubeyko, C. Guyot, ”Method of decreasing write amplification of NAND flash using a journal approach,” U.S. Patent 10,013,346, issued March 7, 2018

    work page 2018

  6. [6]

    V . A. Dubeyko, C. Guyot, ”Method of improving garbage collection efficiency of flash-oriented file sys- tems using a journaling approach,” U.S. Patent Appli- cation US20170139825, published May 18, 2017

    work page 2017

  7. [7]

    V . A. Dubeyko, ”Bitmap Processing for Log-Structured Data Store,” U.S. Patent Application US20190018601, published January 17, 2019

    work page 2019

  8. [8]

    V . A. Dubeyko, S. Song, ”Non-volatile storage system that reclaims bad blocks,” U.S. Patent 10,223,216, is- sued March 5, 2019

    work page 2019

  9. [9]

    V . A. Dubeyko, S. Song, ”Non-volatile storage sys- tem that reclaims bad blocks,” U.S. Patent Application US20190155703, published May 23, 2019

    work page 2019

  10. [10]

    3 Issue 3, Oct

    Agrawal, et al., ”A Five-Year Study of File-System Metadata,” ACM Transactions on Storage (TOS), vol. 3 Issue 3, Oct. 2007, Article No. 9

    work page 2007

  11. [11]

    Wright, ”A nine year study of file system and storage benchmarking,” Trans

    Avishay Traeger, Erez Zadok, Nikolai Joukov, and Charles P. Wright, ”A nine year study of file system and storage benchmarking,” Trans. Storage 4, 2, Arti- cle 5 (May 2008), 56 pages. 45

    work page 2008

  12. [12]

    59- 70, May 1-4, 1999

    Douceur, et al., ”A Large-Scale Study of File-System Contents,” SIGMETRICS ’99 Proceedings of the 1999 ACM SIGMETRICS international conference on Mea- surement and modeling of computer systems, pp. 59- 70, May 1-4, 1999

    work page 1999

  13. [13]

    Available: https://pdfs.semanticscholar.org/48f8/ 1b9339ec3fcee1cc8031575e6f7b84c57c84.pdf, Accessed on: Jun

    Lucas Tan, Fuyao Zhao, Xu Zhang, ”15712 Advanced Operating and Distributed System Android and iOS Platform Study Final Report,” [Online]. Available: https://pdfs.semanticscholar.org/48f8/ 1b9339ec3fcee1cc8031575e6f7b84c57c84.pdf, Accessed on: Jun. 21, 2019

    work page 2019

  14. [14]

    Arpaci-Dusseau, and Remzi H

    Tyler Harter, Chris Dragga, Michael Vaughn, Andrea C. Arpaci-Dusseau, and Remzi H. Arpaci-Dusseau, ”A file is not a file: understanding the I/O behavior of Apple desktop applications,” In Proceedings of the Twenty-Third ACM Symposium on Operating Systems Principles (SOSP ’11). ACM, New York, NY , USA, 71- 83

    work page

  15. [15]

    A. B. Downey, ”The structural cause of file size dis- tributions,” MASCOTS 2001, Proceedings Ninth Inter- national Symposium on Modeling, Analysis and Sim- ulation of Computer and Telecommunication Systems, Cincinnati, OH, USA, 2001, pp. 361-370

    work page 2001

  16. [16]

    M. I. Ullah, F. Ahsan, I. Ahmad and A. F. M. Ishaq, ”Analysis of file system space utilization patterns in UNIX based volumes,” Proceedings of the IEEE Sym- posium on Emerging Technologies, 2005, Islamabad, 2005, pp. 542-546

    work page 2005

  17. [17]

    Miller, Darrell D

    Tim Gibson, Ethan L. Miller, Darrell D. E. Long, ”Long-term File Activity and Inter-Reference Pat- terns,” [Online]. Available: https://www.ssrc. ucsc.edu/papers/CMG-Gibson-1998.pdf, Ac- cessed on: Jun. 25, 2019

    work page 1998

  18. [18]

    Available: https://www.pdl.cmu

    Yifan Wang, ”A Statistical Study for File Sys- tem Meta Data On High Performance Computing Sites,” [Online]. Available: https://www.pdl.cmu. edu/PDL-FTP/HECStorage/Yifan_Final.pdf, Accessed on: Jun. 25, 2019

    work page 2019

  19. [19]

    Wildani, I

    A. Wildani, I. F. Adams and E. L. Miller, ”Single- Snapshot File System Analysis,” 2013 IEEE 21st Inter- national Symposium on Modelling, Analysis and Sim- ulation of Computer and Telecommunication Systems, San Francisco, CA, 2013, pp. 338-341

    work page 2013

  20. [20]

    S. Hui, Z. Rui, C. Jin, L. Lei, W. Fei and X. C. Sheng, ”Analysis of the File System and Block IO Scheduler for SSD in Performance and Energy Consumption,” 2011 IEEE Asia-Pacific Services Computing Confer- ence, Jeju Island, 2011, pp. 48-55

    work page 2011

  21. [21]

    Parthey and R

    D. Parthey and R. Baumgartl, ”Analyzing Access Timing of Removable Flash Media,” 13th IEEE In- ternational Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA 2007), Daegu, 2007, pp. 510-515

    work page 2007

  22. [22]

    Y . Son, H. Kang, H. Han and H. Y . Yeom, ”An Empir- ical Evaluation of NVM Express SSD,” 2015 Interna- tional Conference on Cloud and Autonomic Comput- ing, Boston, MA, 2015, pp. 275-282

    work page 2015

  23. [23]

    K. Zhou, P. Huang, C. Li and H. Wang, ”An Empirical Study on the Interplay between Filesystems and SSD,” 2012 IEEE Seventh International Conference on Net- working, Architecture, and Storage, Xiamen, Fujian, 2012, pp. 124-133

    work page 2012

  24. [24]

    Olivier, J

    P. Olivier, J. Boukhobza and E. Senn, ”Micro- benchmarking Flash Memory File-System Wear Level- ing and Garbage Collection: A Focus on Initial State Impact,” 2012 IEEE 15th International Conference on Computational Science and Engineering, Nicosia, 2012, pp. 437-444

    work page 2012

  25. [25]

    Olivier, J

    P. Olivier, J. Boukhobza and E. Senn, ”Modeling driver level NAND flash memory I/O performance and power consumption for embedded Linux,” 2013 11th Inter- national Symposium on Programming and Systems (ISPS), Algiers, 2013, pp. 143-152

    work page 2013

  26. [26]

    Wei and D

    Y . Wei and D. Shin, ”NAND flash storage device per- formance in Linux file system,” 2011 6th International Conference on Computer Sciences and Convergence Information Technology (ICCIT), Seogwipo, 2011, pp. 574-577

    work page 2011

  27. [27]

    Kim and D

    G. Kim and D. Shin, ”Performance analysis of SSD write using TRIM in NTFS and EXT4,” 2011 6th Inter- national Conference on Computer Sciences and Con- vergence Information Technology (ICCIT), Seogwipo, 2011, pp. 422-423

    work page 2011

  28. [28]

    Park and K

    S. Park and K. Shen, ”A performance evaluation of scientific I/O workloads on Flash-based SSDs,” 2009 IEEE International Conference on Cluster Computing and Workshops, New Orleans, LA, 2009, pp. 1-5

    work page 2009

  29. [29]

    B. Gu, J. Lee, B. M. Jung, J. Seo and H. Shin, ”Uti- lization analysis of trim-enabled NAND flash mem- ory,” 2013 IEEE International Conference on Con- sumer Electronics (ICCE), Las Vegas, NV , 2013, pp. 645-646

    work page 2013

  30. [30]

    Y . Wang, K. Goda, M. Nakano and M. Kitsuregawa, ”Early Experience and Evaluation of File Systems on SSD with Database Applications,” 2010 IEEE Fifth In- ternational Conference on Networking, Architecture, and Storage, Macau, 2010, pp. 467-476. 46

    work page 2010

  31. [31]

    S. S. Rizvi and T. Chung, ”Flash memory SSD based DBMS for high performance computing embedded and multimedia systems,” The 2010 International Con- ference on Computer Engineering & Systems, Cairo, 2010, pp. 183-188

    work page 2010

  32. [32]

    Lin and X

    L. Lin and X. Lizhen, ”The Research of Key Tech- nology in Flash-Based DBMS,” 2009 Sixth Web Infor- mation Systems and Applications Conference, Xuzhou, Jiangsu, 2009, pp. 15-18

    work page 2009

  33. [33]

    J. Chen, J. Wang, Z. Tan and C. Xie, ”Effects of Recursive Update in Copy-on-Write File Systems: A BTRFS Case Study,” in Canadian Journal of Electrical and Computer Engineering, vol. 37, no. 2, pp. 113-122, Spring 2014

    work page 2014

  34. [34]

    Ousterhout, ”The de- sign and implementation of a log-structured file sys- tem,” ACM Trans

    Mendel Rosenblum and John K. Ousterhout, ”The de- sign and implementation of a log-structured file sys- tem,” ACM Trans. Comput. Syst. 10, 1 (February 1992), 26-52

    work page 1992

  35. [35]

    Available: http: //citeseerx.ist.psu.edu/viewdoc/summary? doi=10.1.1.630.3461, Accessed on: Jun

    David Woodhouse, ”JFFS: the journalling flash file system,” [Online]. Available: http: //citeseerx.ist.psu.edu/viewdoc/summary? doi=10.1.1.630.3461, Accessed on: Jun. 20, 2019

    work page 2019

  36. [36]

    Bityutskiy, ”JFFS3 design issues,” [On- line]

    Artem B. Bityutskiy, ”JFFS3 design issues,” [On- line]. Available: http://citeseerx.ist.psu. edu/viewdoc/summary?doi=10.1.1.107.9834, Accessed on: Jun. 20, 2019

    work page 2019

  37. [37]

    Available: http://www.linux-mtd.infradead.org/doc/ ubifs_whitepaper.pdf, Accessed on: Jun

    Adrian Hunter, ”A Brief Introduction to the Design of UBIFS,” [Online]. Available: http://www.linux-mtd.infradead.org/doc/ ubifs_whitepaper.pdf, Accessed on: Jun. 20, 2019

    work page 2019

  38. [38]

    Bityutskiy, ”UBIFS file sys- tem,” [Online]

    Adrian Hunter, Artem B. Bityutskiy, ”UBIFS file sys- tem,” [Online]. Available: http://www.linux-mtd. infradead.org/doc/ubifs.pdf, Accessed on: Jun. 20, 2019

    work page 2019

  39. [39]

    Available: https://yaffs.net/sites/yaffs

    Charles Manning, ”How Y AFFS Works,” [Online]. Available: https://yaffs.net/sites/yaffs. net/files/HowYaffsWorks.pdf, Accessed on: Jun. 20, 2019

    work page 2019

  40. [40]

    [On- line]

    Technical note, the Nilfs version 1: overview. [On- line]. Available: https://nilfs.sourceforge.io/ papers/overview-v1.pdf, Accessed on: Jun. 20, 2019

    work page 2019

  41. [41]

    [Online]

    Ryusuke Konishi, ”Development of a New Log- structured File System for Linux,” Technical Note, Oct 2005. [Online]. Available: https://nilfs. sourceforge.io/papers/nilfs-051019.pdf, Accessed on: Jun. 20, 2019

    work page 2005

  42. [42]

    Available: https://www.researchgate.net/publication/ 228865441_LogFS-finally_a_scalable_flash_ file_system, Accessed on: Jun

    J ¨orn Engel, Robert Mertens, ”LogFS-finally a scalable flash file system,” [Online]. Available: https://www.researchgate.net/publication/ 228865441_LogFS-finally_a_scalable_flash_ file_system, Accessed on: Jun. 21, 2019

    work page 2019

  43. [43]

    USENIX Association, Berkeley, CA, USA, 273-286

    Changman Lee, Dongho Sim, Joo-Young Hwang, and Sangyeun Cho, ”F2FS: a new file system for flash stor- age,” In Proceedings of the 13th USENIX Conference on File and Storage Technologies (FAST’15). USENIX Association, Berkeley, CA, USA, 273-286

    work page

  44. [44]

    Available: http: //citeseerx.ist.psu.edu/viewdoc/download? doi=10.1.1.554.8864&rep=rep1&type=pdf, Accessed on: Jun

    TaeHoon Kim, KwangMu Shin, TaeHoon Lee, KiDong Jung, ”Design of a Reliable NAND Flash Software for Mobile Device,” [Online]. Available: http: //citeseerx.ist.psu.edu/viewdoc/download? doi=10.1.1.554.8864&rep=rep1&type=pdf, Accessed on: Jun. 24, 2019

    work page 2019

  45. [45]

    1241-1244

    Jeong-Ki Kim, Hyung-Seok Lee and Heung-Nam Kim, ”Dual Journaling Store Method for Embedded Sys- tems,” 2006 8th International Conference Advanced Communication Technology, Phoenix Park, 2006, pp. 1241-1244

    work page 2006

  46. [46]

    S. O. Park and S. J. Kim, ”An Efficient Array File Sys- tem for Multiple Small-Capacity NAND Flash Memo- ries,” 2011 14th International Conference on Network- Based Information Systems, Tirana, 2011, pp. 569-572

    work page 2011

  47. [47]

    J. Kim, H. Jo, H. Shim, J. Kim and S. Maeng, ”Effi- cient Metadata Management for Flash File Systems,” 2008 11th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Com- puting (ISORC), Orlando, FL, 2008, pp. 535-540

    work page 2008

  48. [48]

    S. O. Park and S. J. Kim, ”An efficient multimedia file system for NAND flash memory storage,” in IEEE Transactions on Consumer Electronics, vol. 55, no. 1, pp. 139-145, February 2009

    work page 2009

  49. [49]

    Seung-Ho Lim and Kyu-Ho Park, ”An efficient NAND flash file system for flash memory storage,” in IEEE Transactions on Computers, vol. 55, no. 7, pp. 906-912, July 2006

    work page 2006

  50. [50]

    H. Kim, Y . Won and S. Kang, ”Embedded NAND flash file system for mobile multimedia devices,” in IEEE Transactions on Consumer Electronics, vol. 55, no. 2, pp. 545-552, May 2009

    work page 2009

  51. [51]

    C. T. Chen, C. H. Chen and W. T. Huang, ”Energy- aware management of NAND type flash file system,” in Electronics Letters, vol. 42, no. 14, pp. 795-796, 6 July 2006. 47

    work page 2006

  52. [52]

    A. S. Ramasamy and P. Karantharaj, ”File system and storage array design challenges for flash mem- ory,” 2014 International Conference on Green Comput- ing Communication and Electrical Engineering (ICGC- CEE), Coimbatore, 2014, pp. 1-8

    work page 2014

  53. [53]

    Nahill and Z

    B. Nahill and Z. Zilic, ”FLogFS: A lightweight flash log file system,” 2015 IEEE 12th International Confer- ence on Wearable and Implantable Body Sensor Net- works (BSN), Cambridge, MA, 2015, pp. 1-6

    work page 2015

  54. [54]

    Yang Ou, Xiaoquan Wu, Nong Xiao, Fang Liu and Wei Chen, ”HIFFS: A Hybrid Index for Flash File System,” 2015 IEEE International Conference on Networking, Architecture and Storage (NAS), Boston, MA, 2015, pp. 363-364

    work page 2015

  55. [55]

    Huang, G

    P. Huang, G. Wan, K. Zhou, M. Huang, C. Li and H. Wang, ”Improve Effective Capacity and Lifetime of Solid State Drives,” 2013 IEEE Eighth International Conference on Networking, Architecture and Storage, Xi’an, 2013, pp. 50-59

    work page 2013

  56. [56]

    Yang and C

    S. Yang and C. Wu, ”A Low-Memory Management for Log-Based File Systems on Flash Memory,” 2009 15th IEEE International Conference on Embedded and Real- Time Computing Systems and Applications, Beijing, 2009, pp. 219-227

    work page 2009

  57. [57]

    W. Qiu, X. Chen, N. Xiao, F. Liu and Z. Chen, ”A New Exploration to Build Flash-Based Storage Sys- tems by Co-designing File System and FTL,” 2013 IEEE 16th International Conference on Computational Science and Engineering, Sydney, NSW, 2013, pp. 925-932

    work page 2013

  58. [58]

    T. Chen, X. Wang, W. Hu and W. Duan, ”A New Type of NAND Flash-Based File System: Design and Imple- mentation,” 2006 International Conference on Wireless Communications, Networking and Mobile Computing, Wuhan, 2006, pp. 1-4

    work page 2006

  59. [59]

    S. Lee, J. Kim and A. Mithal, ”Refactored Design of I/O Architecture for Flash Storage,” in IEEE Computer Architecture Letters, vol. 14, no. 1, pp. 70-74, 1 Jan.- June 2015

    work page 2015

  60. [60]

    Junkil Ryu and C. Park, ”A technique to enhance per- formance of log-based file systems for flash memory in embedded systems,” 2007 2nd International Confer- ence on Digital Information Management, Lyon, 2007, pp. 580-582

    work page 2007

  61. [61]

    Byungjo Kim, Dong Hyun Kang, Changwoo Min and Young Ik Eom, ”Understanding implications of trim, discard, and background command for eMMC storage device,” 2014 IEEE 3rd Global Conference on Con- sumer Electronics (GCCE), Tokyo, 2014, pp. 709-710

    work page 2014

  62. [62]

    C. Min, S. Lee and Y . I. Eom, ”Design and Implemen- tation of a Log-Structured File System for Flash-Based Solid State Drives,” in IEEE Transactions on Comput- ers, vol. 63, no. 9, pp. 2215-2227, Sept. 2014

    work page 2014

  63. [63]

    Jun Wang and Yiming Hu, ”A novel reordering write buffer to improve write performance of log-structured file systems,” in IEEE Transactions on Computers, vol. 52, no. 12, pp. 1559-1572, Dec. 2003

    work page 2003

  64. [64]

    285- 292

    Jun Wang and Yiming Hu, ”PROFS-performance- oriented data reorganization for log-structured file sys- tem on multi-zone disks,” MASCOTS 2001, Proceed- ings Ninth International Symposium on Modeling, Analysis and Simulation of Computer and Telecommu- nication Systems, Cincinnati, OH, USA, 2001, pp. 285- 292

    work page 2001

  65. [65]

    Agarwal and M

    R. Agarwal and M. Marrow, ”A closed-form expression for write amplification in NAND Flash,” 2010 IEEE Globecom Workshops, Miami, FL, 2010, pp. 1846- 1850

    work page 2010

  66. [66]

    Jagmohan, M

    A. Jagmohan, M. Franceschini and L. Lastras, ”Write amplification reduction in NAND Flash through multi- write coding,” 2010 IEEE 26th Symposium on Mass Storage Systems and Technologies (MSST), Incline Village, NV , 2010, pp. 1-6

    work page 2010

  67. [67]

    Chang and T

    Y . Chang and T. Kuo, ”A commitment-based man- agement strategy for the performance and reliability enhancement of flash-memory storage systems,” 2009 46th ACM/IEEE Design Automation Conference, San Francisco, CA, 2009, pp. 858-863

    work page 2009

  68. [68]

    Tei-Wei Kuo, Jen-Wei Hsieh, Li-Pin Chang and Yuan- Hao Chang, ”Configurability of performance and over- heads in flash management,” Asia and South Pacific Conference on Design Automation, 2006., Yokohama, 2006, p. 8

    work page 2006

  69. [69]

    Hsieh, C

    J. Hsieh, C. Wu and G. Chiu, ”Design and Imple- mentation for Multi-level Cell Flash Memory Storage Systems,” 2010 IEEE 16th International Conference on Embedded and Real-Time Computing Systems and Applications, Macau SAR, 2010, pp. 247-252

    work page 2010

  70. [70]

    C. Park, W. Cheon, Y . Lee, M. Jung, W. Cho and H. Yoon, ”A Re-configurable FTL (Flash Translation Layer) Architecture for NAND Flash based Appli- cations,” 18th IEEE/IFIP International Workshop on Rapid System Prototyping (RSP ’07), Porto Alegre, 2007, pp. 202-208

    work page 2007

  71. [71]

    J. Lee, H. Kim, H. Kim, J. Park and M. Ryu, ”A se- quentializing device driver for optimizing random write 48 performance of eSSD,” 2014 IEEE International Con- ference on Consumer Electronics (ICCE), Las Vegas, NV , 2014, pp. 432-433

    work page 2014

  72. [72]

    Y . He, S. Wan, N. Xiong and J. H. Park, ”A New Prefetching Strategy Based on Access Density in Linux,” International Symposium on Computer Sci- ence and its Applications, Hobart, ACT, 2008, pp. 22- 27

    work page 2008

  73. [73]

    Dingqing Hu, Changsheng Xie and C. CaiBin, ”A Study of Parallel Prefetching Algorithms Using Trace- Driven Simulation,” Sixth International Conference on Parallel and Distributed Computing Applications and Technologies (PDCAT’05), Dalian, China, 2005, pp. 476-478

    work page 2005

  74. [74]

    Y . Kang, J. Yang and E. L. Miller, ”Efficient Storage Management for Object-based Flash Memory,” 2010 IEEE International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems, Miami Beach, FL, 2010, pp. 407-409

    work page 2010

  75. [75]

    Q. Xie et al, ”Research on the Framework of NAND FLASH Based Object-Based-Storage-Device,” 2012 Second International Conference on Intelligent System Design and Engineering Application, Sanya, Hainan, 2012, pp. 1298-1301

    work page 2012

  76. [76]

    Available: http://citeseerx.ist.psu.edu/ viewdoc/download?doi=10.1.1.219.7269&rep= rep1&type=pdf, Accessed on: Jun

    Goetz Graefe, ”Modern B-Tree Techniques,” [Online]. Available: http://citeseerx.ist.psu.edu/ viewdoc/download?doi=10.1.1.219.7269&rep= rep1&type=pdf, Accessed on: Jun. 21, 2019

    work page 2019

  77. [77]

    J. Ahn, D. Kang, D. Jung, J. Kim and S. Maeng, ” μ* -Tree: An Ordered Index Structure for NAND Flash Memory with Adaptive Page Layout Scheme,” in IEEE Transactions on Computers, vol. 62, no. 4, pp. 784-797, April 2013

    work page 2013

  78. [78]

    Lee and S

    C. Lee and S. Lim, ”Caching and Deferred Write of Metadata for Yaffs2 Flash File System,” 2011 IFIP 9th International Conference on Embedded and Ubiquitous Computing, Melbourne, VIC, 2011, pp. 41-46

    work page 2011

  79. [79]

    He et al., ”Discovering Structure in Unstructured I/O,” 2012 SC Companion: High Performance Com- puting, Networking Storage and Analysis, Salt Lake City, UT, 2012, pp

    J. He et al., ”Discovering Structure in Unstructured I/O,” 2012 SC Companion: High Performance Com- puting, Networking Storage and Analysis, Salt Lake City, UT, 2012, pp. 1-6

    work page 2012

  80. [80]

    Arul, Jia-Shian Wu, I

    Tsozen Yeh, J. Arul, Jia-Shian Wu, I. -. Chen and Kuo- Hsin Tan, ”Using File Grouping to Improve the Disk Performance (Extended Abstract),” 2006 15th IEEE International Conference on High Performance Dis- tributed Computing, Paris, 2006, pp. 365-366

    work page 2006

Showing first 80 references.