Home Messages Index
[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index

[News] Linux File System Throughput Benchmarks

  • Subject: [News] Linux File System Throughput Benchmarks
  • From: Roy Schestowitz <newsgroups@xxxxxxxxxxxxxxx>
  • Date: Wed, 16 Sep 2009 10:55:48 +0100
  • Followup-to: comp.os.linux.advocacy
  • Newsgroups: comp.os.linux.advocacy
  • User-agent: KNode/4.3.1
Hash: SHA1

I Feel the Need for Speed: Linux File System Throughput Performance, Part 1

,----[ Quote ]
| In two previous articles (here and here) we explored the 
| metadata performance of a number of Linux file systems using a 
| single micro-benchmark: fdtree.
| fdtree as a micro-benchmark is very attractive because it is a 
| simple bash script that uses recursion, forcing all cores to be 
| used (extremely important with modern processors). It tests the 
| ability of the file system to simply create directories and 
| files in a tree-structure. The file systems tested typically 
| used their default options (except for ext3 and ext4) so tuning 
| the file systems for this specific benchmark was not tested. 



Comparison of File Systems And Speeding Up Applications

,----[ Quote ]
| The results are very impressive and the final patches to libvirt were
| finalised pretty quickly. They're now in the development branch libvirt.
| Coming soon to a virtual machine management application near you.


BFS Scheduler Benchmarks

,----[ Quote ]
| While BFS lost with the Threaded I/O Tester disk benchmark, it won with the
| PostMark test profile. The BFS scheduler delivered just fewer than 5% more
| transactions per second than the Completely Fair Scheduler.
| Nine tests carried out on a single low-end system (we may carry out more tests
| on multiple systems if there is sufficient interest) is not enough to call one
| scheduler better than the other.



Finding the Fastest Filesystem

,----[ Quote ]
| I was amazed by the numbers. For all but one uniform subset of tests, XFS was
| the clear winner. I even tried synchronous mounts and encrypted volumes as
| little "what if?" exercises, and XFS still came out on top. The parameter
| that cost XFS a total victory was the CFQ elevator on a slower system; ext3
| won most of those cases.


Serverwide Performance Benchmarking

,----[ Quote ]
| Based on my benchmark numbers I think there are a few statements that can be
| labeled "conclusions":
|     * Tuning your database performance parameters is the best way to improve
|        overall system throughput on a shared database and webserver.
|     * Except for the data-integrity focused filesystems like Ext3 in
|       journalled or ordered mode, filesystem performance is pretty much not a
|       major factor contributing to database and webserver throughput.  
|     * Enabling noatime does not make a big (over 5%) impact on filesystem
|       performance for typical database and webserver loads.
|     * APC can give a measurable performance boost in PHP script processing,
|       of course largely depending on the size of your scripts.


Linux Cure for Exchange Storage Bloat

,----[ Quote ]
| Users spend precious work hours copying e-mails to desktop PST files â which
| may be backed up rarely or never â or other secondary e-mail storage systems,
| copying them to their hard drives, or just deleting these important business
| records. If an employee needs to reference an old e-mail, he has to remember
| what it was called and where he filed it, then reload the PST file and search
| for the message. The productivity hit can be huge, and there is no guarantee
| the employee will actually find the critical information he is looking for.      
| [...]
| Modern Linux filing systems such as XFS and Ext3 leverage these optimizations
| and support features such as journaling (to ensure hierarchy integrity
| following a power cut), clustering and replication (DRBD), and snapshots (via
| LVM). With modern e-mail server architecture, IT can complete backup
| operations without requiring a freeze or a database snapshot.    


Delve deep into drives

,----[ Quote ]
| Prabhakaran then goes on tracking bugs in all UNIX drivers, describing
| inconsistencies and danger points, tracing the outline of a tougher file
| system and then describing how to create an evolved file system that would
| unite the advantages of most current file systems and overcome most of their
| shortcomings. At the same time, he tries to describe how moving critical
| logical pieces from the driver to the kernel (and therefore sharing this code
| from one file system driver to the other) may make development easier and
| faster while at the same time strengthening existing FS.      


ZFS, XFS, EXT4 Filesystems Compared

,----[ Quote ]
| EXT4 is fast for metadata operations, tar, untar, cpio, and postmark. EXT4 is
| much faster than the others under FFSB. EXT4 with hardware RAID and external
| journal device is ludicrously fast. EXT4 seems to have a bad interaction with
| software RAID, probably because mkfs fails to query the RAID layout when
| setting the filesystem parameters.    
| ZFS has excellent performance on metadata tests. ZFS has very bad sequential
| transfer with hardware RAID and appalling sequential transfer with software
| RAID. ZFS can copy the linux kernel source code in only 3 seconds! ZFS has
| equal latency for read and write requests under mixed loads, which is good.  
| XFS has good sequential transfer under Bonnie++. Oddly XFS has better
| sequential reads when using an external journal, which makes little sense. Is  
| noatime broken on XFS? XFS is very slow on all the metadata tests. XFS takes
| the RAID layout into consideration and it performs well on randomio with
| hardware or software RAID.    


OpenSolaris ZFS vs. Linux ext3 RAID5

,----[ Quote ]
| Few overarching conclusions can be drawn from the limited results of this
| study. Certainly, there are situations in which the Solaris/RAID-Z
| configuration appears to outperform the Ubuntu/RAID-5 configuration. Many
| questions remain regarding the large discrepancy in CPU usage for small-file
| operations. Likewise, the Ubuntu/RAID-5 configuration appears to perform
| slightly better in certain situations, though not overwhelmingly so. At best,
| under these default configurations, one can say that overall the Solaris
| configuration performs no worse, and indicates that it might perform better
| under live operating conditions. The latter, though, is largely speculation.


Why so many filesystems for Linux? What's the difference?

,----[ Quote ]
|     * EXT3      
|         * Most popular Linux file system, limited scalability in size and
|         number of files      
|         * Journaled      
|         * POSIX extended access control
|     EXT3
|     file system is a journaled file system that has the greatest use in
|     Linux today. It is the "Linux" File system. It is quite robust and
|     quick, although it does not scale well to large volumes nor a great
|     number of files. Recently a scalability feature was added called
|     htrees, which significantly improved EXT3's scalability.
| [...]
|     * FAT32      
|         * Most limited file system, but most ubiquitous      
|         * Not Journaled      
|         * No access controls
|     FAT32
|     is the crudest of the file systems listed. Its popularity is with its
|     widespread use and popularity in the Windows desktop world and that it
|     has made its way into being the file system in flash RAM devices
|     (digital cameras, USB memory sticks, etc.). It has no built in security
|     access control, so is small and works well in these portable and
|     embedded applications. It scales the least of the file systems listed.
|     Most systems have FAT32 compatibility support due to its ubiquity.

Version: GnuPG v1.4.9 (GNU/Linux)


[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index