Increased Performance of ext4 vs. I also have a separate zfs pool for either additional storage or VMs running on zfs (for snapshots). Xfs is the default for redhat. As always, your mileage may vary 🙂. which btw you should put in here then as well. But I was more talking to the XFS vs EXT4 comparison. my nextcloud site). IOSTAT also showing EXT4 was at 98. Here are some alternatives: XFS. After earlier in the week delivering solid-state drive file-system benchmarks in comparing the Linux 3. When running one copy of the SQLite embedded database library, the XFS file-system had a slim lead over NILFS2 and F2FS while EXT4 was the slowest on this Linux 5. Or when it came to testing the single Seagate IronWolf 6TB HDD performance, Btrfs and EXT4 were performing about the same with. It is a rock-solid option since it has been around for long, bringing with it all the years of. Already have an account? Sign in to comment. The observation was that XFS is useful when your machine has multiple cores and fast disk that XFS can utilize. However benchmarks test quite narrow parameters which may not be reflected by running an OS. Phoronix: Linux 4. It is suitable for PC platforms and network. I chose two established journaling filesystems EXT4 and XFS two modern Copy on write systems that also feature inline compression ZFS and BTRFS and as a relative benchmark for the achievable compression SquashFS. Although XFS is good, in practice I've found ext4 to be slightly faster. XFS vs. For large block sizes, such as 64KiB, both filesystems are on par. 2070 tps). xfs -l size=64m (notes fromHas anyone compared the IO performance of WSL2 "emulated disk" vs a bare physical disk from wsl --mount --bare ? (Is there any comparison of ext4 vs XFS vs ZFS? I will run fio myself but I'd like to compare benchmarks to avoid wasting too much time). the COW which saves alot of space and increases the speed. Observations. 7. The system was set for Performance; whatever energy saving features I could find in the BIOS were turned off. Whether for. After a week of testing Btrfs on my laptop, I can conclude that there is a noticeable performance penalty vs Ext4 or XFS. XFS, EXT4, and BTRFS are file systems commonly used in Linux-based operating systems. With Dbench as well, XFS sees the largest drop in performance from KPTI and Retpoline support. Ext4 is an open-source, enhanced filesystem for Linux OSs that supersedes ext3 in terms of speed, dependability, and expansiveness. While looking at the filesystem options it seems like BTRFS is a lot more stable than it was the last time I had to install arch so now I am seriously considering using it. Various benchmarks have concluded that the actual ext4 file system can perform a variety of read-write operations faster than an NTFS partition. 3. ZFS is a single file system that creates sub-volumes when needed. However, LVM can provide great performance as well, especially when used with specific (good-performing) filesystems like XFS or Ext4. See Swap#Performance. Snapshots, transparent compression and quite importantly blocklevel checksums. file-system comparison, here are some fresh benchmarks looking at the Btrfs, EXT4, F2FS, and. Compared to ext4, XFS has unlimited inode allocation, advanced allocation hinting (if you need it) and, in recent version, reflink support (but they need to be explicitly enabled in Ubuntu 18. Neither file system consistently outperforms the other in all workloads. The four hard drives used for testing were 6TB Seagate IronWolf NAS (ST6000VN0033. – in the case of SATA/SSD, the ext4 scalability issue has an impact on tps rate after 256 threads and drop is 10-15%. XFS vs. The problem (which i understand is fairly common) is that performance of a single NVMe drive on zfs vs ext4 is atrocious. This enables extreme scalability of IO threads, filesystem bandwidth, file and filesystem size when spanning multiple storage devices. XFS is a high-performance, journaling file system designed for high scalability. g. EXT4 is the successor of EXT3, the most used Linux file system. 14 stable. Now there are a few others that are really interesting for SSD/NVMe, such as F2FS, XFS, etc. The benchmarks in this article are looking at the EXT4 / Btrfs / XFS / F2FS file-systems under the Linux 4. 0 while today is just a comparison of six file-systems using a traditional HDD. Small_Light_9964 • 1 yr. XFS performance there for flash storage where this file-system is designed. Additionally, XFS supports standard SSD. À partir de Red Hat Enterprise Linux 7. ZFS allows users to move these files anywhere and even to attach them to the ZFS on. XFS still has some reliability issues, but could be good for a large data store where speed matters but rare data loss (e. Linux EXT4/Btrfs RAID. Back when Bcachefs debuted in. Each of these file systems has its own way of organizing data, merits, and demerits. Various benchmarks have concluded that the actual ext4 file system can perform a variety of read-write operations faster than an NTFS partition. Hi folks, just wondering if anyone has experience with running clickhouse on ext4 vs xfs? And if there is any benchmark of ext4 vs xfs for clickhouse data volume? Specifically with high IOPS. Exfat compatibility is excellent (read and write) with Apple AND Microsoft AND Linux. This is addressed in this knowledge base article; the main consideration for you will be the support levels available: Ext4 is supported up to 50TB, XFS up to 500TB. 对于一些文件系统如Ext4等，在硬盘格式化时就全部确定了，而对于XFS则是动态生成的，BtrfS则是更特别的动态实现。. With not having the time to conduct the usual kernel version vs. ext4 is the successor to ext3. Ext4 seems better suited for lower-spec configurations although it will work just fine on faster ones as well, and performance-wise still better than btrfs in most cases. ext4. 5. ext3/ext4: Use the barrier=0 mount option to disable barriers. • Specification defines an optimized register interface, command set and feature set. User quotas for each shared folder. The good news is that both ext4 and XFS facilitate excellent performance for database systems. XFS (2002) – originally SGI Irix 5. 0 File-System Benchmarks: Btrfs vs. Linux 4. With 4K random reads by FIO, the SATA/USB performance was flat across. XFS . However, unlike Extended 4, it is not possible to disable journaling, thus it can be iffy to use on an SSD. We benchmarked XFS vs EXT4 file system on these storage devices as well. at least thin-LVM as storage type is something that people might use to provide the guests. Though not as large of a difference when comparing to an SD card. In the case of the Intel 900p SSD, the XFS results were too fast to accurately measure while EXT4 and F2FS took just two seconds to complete while Btrfs took six seconds. xfs: 0. File systems may be resized after creation, with certain limitations. Although XFS is good, in practice I've found ext4 to be slightly faster. After deciding to use LVM2 as volumemanager on our servers there was also the wish for an online resizeable filesystem. The way you describe this workload, I think it is not very demanding. Depending on the space in question, I typically end up using both ext4 (on lvm/mdadm) and zfs (directly over raw disks). To make the benchmarks above more clear, it might might help to normalise them relative to the performance of ext4 on each disk: ops randappend SMR. Ext4 is the evolution of the most used Linux filesystem, Ext3. If you end up increasing the size of the box then it's going to become more relevant. Last week I posted some fresh Linux file-system tests on a hard drive but for those preferring solid-state drives, here are some fresh benchmarks. The benchmark results of three most common file systems under Linux environment were given in this paper. If you want raw speed, XFS is king. Main features: Data protection features, including snapshot, replication, and point-in-time recovery. I used to format XFS using mkfs. If we apply a fix by mounting ext4 with dioread_nolock or use xfs, throughput looks good. On an ssd desktop you will NOT notice a difference in performance between ext4 and xfs. EXT4, XFS and ZFS comparison. Ext4 focuses on high-performance and scalability. というのをベースにするとXFSが良い。 一般的にlinuxのブロックサイズは4kなので、xfsのほうが良さそう。 MySQLでページサイズ大きめならext4でもよい。xfsだとブロックサイズが大きくなるにつれて遅くなってる傾向が見える。ext4. For bare metal mail server I'd go ZFS all the way tho. At 32 threads ext4 was 28% faster (2345 tps vs. btrfs: 1. The ext4 is an old file system that is the default in several Linux distributions, such as Ubuntu. 0, 82. Still, the filesystem is constantly called “high performance,” meaning it. XFS supports larger file sizes and. When XFS was designed, “high performance” meant a. ZFS On Linux Benchmarks Storage : 2019-01-26: FreeBSD ZFS vs. Refer to corresponding file system page in case there were performance improvements instructions, e. XFS is a robust and mature 64-bit journaling file system that supports very large files (scales to exabytes) and file systems on a single host. It seems that the new file system may be applied more. EXT4 vs. 4 HDD RAID performance per his request with Btrfs, EXT4, and XFS while using consumer HDDs and an AMD Ryzen APU setup that could work out for a NAS type low-power system for anyone else that may be interested. Through many years of development, it is one of the most stable file systems. For single disks over 4T, I would consider xfs over zfs or ext4. 1. I've seen that EXT4 has better random I/O performance than XFS, especially on small reads and writes. F2FS vs. XFS vs. Given Canonical has brought. ext4 to specify a file system label. ZFS meanwhile still handily beat out the UFS competition -- the Sun/Oracle ZFS was 53% faster than UFS+S and an impressive 2. I've read and have anecdotally (not scientific and could be affected by other things) experienced Btrfs being slower than ext4. The next subsections detail read workloads, write workloads, meta-data workloads, macro workloads, and the impact of performance vs. Both VM’s are on a XFS based filesystem on the hypervisor. 4. El sistema de archivos es mayor de 2 TiB con inodos de 512 bytes. But time is going, and the. Edit: fsdump / fsrestore means the corresponding system backup and restore to for that file system. ZFS has built-in RAID support with various RAID-Z levels (RAID-Z, RAID-Z2, and RAID-Z3). BTRFS is basically the Linux version of ZFS (rather than just ZFS ported to Linux), but it still needs work around RAID. 6. As a DBA, this is what you want to see on your systems—minimum differences (jitter) during the whole benchmark run. With Bcachefs on its trek towards the mainline Linux kernel, this week I conducted some benchmarks using the very latest Bcachefs file-system code and compared its performance to the mainline Btrfs, EXT4, F2FS, and XFS file-system competitors on both rotating and solid-state storage. Beyond just testing the EXT4, Btrfs, and NILFS2 file-systems, we also threw in some results from EXT3 and XFS. XFS uses the copy of the update for journal commit while EXT4 uses the original page cache entry for journal com-mit. XFS allows multi-threaded concurrent journal commit while EXT4 has single threaded serial commit. NTFS. With the same benchmark, very favorable to XFS, I added a ZFS L2ARC and that completely reversed the situation, more than tripling the ZFS results,. 19 and Linux 4. Pros: Individual file size: 16GB to 2TB. XFS provides a more efficient data organization system with higher performance capabilities but less reliability than ZFS, which offers improved accessibility as well as greater levels of data integrity. 6. The charts show sequential reads (top) and writes (bottom) on XFS (left) and EXT4. Sign up for free to join this conversation on GitHub . The EXT4 f ile system is 48-bit with a maximum file size of 1 exbibyte, depending on the host operating system. Another interesting result is that XFS seems to have improved on SSDs between kernels 3. XFS File System. I ran performance benchmarks comparing XFS with EXT4 for MongoDB on AWS EC2 to find out exactly what you were wondering about. . For anything with higher capability, XFS tends to be faster. EXT3, EXT4, XFS EXT3 (2001) / EXT4 (2008) – evolution of original Linux file system (ext, ext2,. I've seen benchmarks (eg: this one) that put btrfs considerably slower than ext4. ext4 has better performance with large files. It's a mature filesystem and offers online defragmentation and can. File-systems tested on the NVMe SSD included Btrfs, EXT4, F2FS, XFS, and NTFS. I used a Dell R630 machine with two E5-2699 CPUs in it. Published very recently by Phoronix, a series of benchmark tests. The XFS is a high-performance 64-bit journaling file system. I use Warp and mc support perf for benchmark. Linux 5. 2. The one they your distribution recommends. This is the first time that the new EXT4 and Btrfs and NILFS2 filesystems have been directly compared when it comes to their disk performance though the results may surprise. Since then, however, ZFS on Linux has progressed a lot and I also learned how to better tune it. Ext4 is the default file system on most Linux distributions for a reason. The result is a filesystem with an improved. This page is powered by a knowledgeable community that helps you make an informed decision. Also, I found out the sysbench benchmark I used at the time was not a fair choice since the dataset it generates compresses much less than a. Honestly I wasn't aware of the huge amount of extends still created - that explains a bit. XFS also tended to perform well along with the seldom mentioned NILFS2. The hard drive used for testing in this article was the Western Digital VelociRaptor. On a slow Linux box with an ext4 filesystem, the same operation takes less than a second. ext3 is the most common format. CoW filesystems like BtrFS are great and full of advantages, but the performance drop away from XFS is notable. That's disgusting enough for me not to want it. Across the three tested RAID modes, EXT4 was performing the worst. To achieve expected performance by tweaking the IRQ affinity, consider few important parameters like Linux handling of the server topology, NIC driver stack, default. there were many tentatives to bring XFS on front, but, again, historically, there were always some issues as soon as workload became IO-bound. Generally NAS server operating systems like QNAP, Asustor or Synology. But btrfs also aims to provide next-gen features that break the. XFS tends to perform better for systems that run on higher capacity. The server I'm working with is:2. 현재 Ext4는 Red Hat Enterprise Linux 6의 기본 파일 시스템으로 단일 파일. exFAT vs NTFS. Presently, Ext4 is the maintainer deployed in the Android OS. NT-based Windows did not have any support for FAT32 up to. For large sequential reads and writes XFS is a little bit better. With a decent CPU transparent compression can even improve the performance. Unfortunately Synology uses ext4 and btrfs; no support for xfs out of the box. A filesystem is ext4 if it uses a feature that isn't in the ext3 driver, and ext3 if it isn't ext4 but uses a feature that isn't in the ext2 driver. Posted by Dimitri Kravtchuk on Wed 13 May 2020 20:15 UTC Tags: innodb, Benchmarks, xfs, ext4, MySQL, Performance, DoubleWrite. The most commonly used are Ext4, Btrfs, XFS, and ZFS which is the most recent file system released back in 2018. Operating system: Raw-VM is Ubuntu 12. My biggest issue with any file system other than EXT4 is that a lot of linux programs are built and tested on EXT4. The four hard drives used for testing were 6TB Seagate IronWolf NAS (ST6000VN0033-2EE) hard drives and the. There are not three filesystem formats, but filesystem formats defined by a combination of features. 4% utilization. The Infortrend RAID is a 24-disk box arranged as two RAID-6 arrays of 12 disks each, each disk 1 TB. Not just permissions, but moving them or getting file sizes, too. I think in many ways btrfs is the better filesystem, but I seem to have noticed that it takes longer to copy data than on ext4. XFS was more fragile, but the issue seems to be fixed. 另外，我们常说的file对象，它用于关联进程和dentry对象的. historically with MySQL we always observed better performance and more stable processing on EXT4. With not having the time to conduct the usual kernel version vs. Here are a few other differences: Features: Btrfs has more advanced features, such as snapshots, data integrity checks, and built-in RAID support. For large sequential reads and writes XFS is a little bit better. Ext4#Improving performance and XFS#Performance. XFS is a full 64-bit filesystem and in theory it is capable of handling filesystems as large as 8 Exabytes For Oracle Linux, we support up to 100TB. XFS does not require extensive reading. 24. Btrfs on SSD, XFS on HDD. Large local PCI-E NVMe "scratch" caches on HPC and VFX nodes are exposed via XFS for their incredible performance. but for the shared servers with many users you might consider xfs for the parallel IO and number of files. Overall, except for application launch time, benchmark results show that ZFS is the slowest file system in terms of read and write speed due to its COW operating type, while EXT4 is usually the fastest system. BTRFS is newer, and the performance is not as good in many cases, but it is not far off. If you have single vmdk on dedicated VMFS I wouldn't expect any difference compare to RDM. EXT4 vs. Each of the tested file-systems were carried out with the default mount options in an out-of-the-box manner. I would recommend choosing between ext4 and xfs filesystems. ZFS 101—Understanding ZFS storage and performance. EXT4 lacks more robust features but is stable and well-supported on all Linux operating systems. Recent File System Benchmarks - BTRFS XFX Ext4 F2FS. being written when I compare the traces), when I look at a representative “same” action I see 5 ops on XFS…there are only 2 for the same action on EXT4. One of the primary advantages of ext4 is that it is a journaled file system, meaning that it. org's git. Btrfs' RAID on Linux 5. Small to Medium Enterprises: While ext3 suffices for businesses with modest data needs, scalability visionaries would do well considering ext4. 7 Average speed : 87. Rep: XFS has unbalanced performance, but in the best use case blows away many other formats. XFS was originally developed by Silicon Graphics for IRIX and later ported to Linux. So I recreated the benchmark fs as xfs and repeated the sysbench run. all kinds for nice features (like extents, subsecond timestamps) which ext3 does not have. At the same time, XFS often required a kernel compile, so it got less attention from end. 3. XFS is better in general with WT, as the MongoDB production notes suggest. Btrfs Benchmarks comparison, here is a wider look at mainline file-systems on the Linux 4. It can hold up to 1 billion terabytes of data. 7. Server with complex storage needs including redundancy and you need high uptime, and you have the budget. an XFS filesystem on a straight disk partition. At the same time, XFS often required a kernel compile, so it got less attention from end. Choosing the correct file system to use on a NAS server is a very important decision, depending on the use that we are going to give it, we can choose one file system or another, since it could provide us with higher performance, better data integrity and Other features. Since then, however, ZFS on Linux has progressed a lot and I also learned how to better tune it. In the future, Linux distributions will gradually shift towards BtrFS. Seeking around those files which a DB will do may yield different. EDIT 1: Added that BTRFS is the default filesystem for Red Hat but only on Fedora. A 3TB / volume and the software in /opt routinely chews up disk space. The XFS file system is an extension of the extent file system. Native file systems (e. file-system comparison, here are some fresh benchmarks looking at the Btrfs, EXT4, F2FS, and XFS file-system benchmarks on a speedy WD_BLACK SN850 NVMe solid-state drive. Each of the five file-systems were tested on the same NVM Express SSD from the Linux 4. It provides good performance with SSD and supports the TRIM (and FITRIM) feature to keep good SSD performance over time (this clears unused memory blocks for quick later write access). XFS (2002) – originally SGI Irix 5. I'll have out our usual file-system/kernel comparison out soon from an SSD in looking at Btrfs/XFS/EXT4/F2FS between Linux 3. If Btrfs and EXT4 aren’t cutting it for you or aren’t supported by your choice of distro, there are a few other popular choices for file systems. This can be achieved by various means, including copying data back and. Btrfs was edging ahead of XFS and Btrfs with the IOzone write test although the performance on the Linux 3. An external ext4 disk, mounted by WSL2 as a bare drive is for all intents and purposes a. To be honest, one of the things that comes last in people’s thinking is to look at which file system on their PC is being used. 1. Still, the filesystem is constantly called “high performance,” meaning it makes perfect sense to turn to this filesystem for high performance drives. However, the performance of ZFS on FreeBSD/PC-BSD 8. Snapraid says if the disk size is below 16TB there are no limitations, if above 16TB the parity drive has to be XFS because the parity is a single file and EXT4 has a file size limit of 16TB. Continue readingWindows has always been terribly slow to update, say, all file permissions in a large directory structure. It supports large file systems and provides excellent scalability and reliability. Observations. 9: “ext4: Allow parallel DIO reads”. For the most. It has wider compatibility than NTFS, which means it's more likely to work with media players, consoles, and a variety of. 2) (surprisingly, the loopback benchmark looks better than the raw-disk benchmark, presumably because of the smaller size of the loopback device, thus less time is spent on the actual sync-to-disk) Benchmark setupDependending on the hardware, ext4 will generally have a bit better performance. 2020. Ext4 file system is an ideal choice. With the WiredTiger storage engine, use of XFS is strongly recommended to avoid performance issues that may occur. Various benchmarks have concluded that the actual ext4 file system can perform a variety of read-write operations faster than an NTFS partition. • Main goal of NVMe is to scale performance and standardize the PCIe SSD Interface • NVMe can be used as local storage or as cache for slower storage devices • Nvme performance: – File system: when compared to SAS SSD by 400% – Cache device: when compared to SAS 12Gpbs HDD by 450% (Read/Write) to 4702 % (Read) The XFS file system is an extension of the extent file system. Picking a filesystem is not really relevant on a Desktop computer. Ext4 is also a more traditional file system, while XFS provides more scalability and is better suited for large file systems. It is because XFS consumes double the CPU-per-metadata operation compared to Ext3 and Ext4. 1. Comparison of file archivers. Between EXT4 and XFS which file system is better when an application uses multiple threads to read/write large amount of small files on a SSD. F2FS vs. 4 usage of the XFS file system. As cotas XFS não são uma opção remountable. I tested an XFS filesystem on an LVM physical volume vs. If you buy a modern drive, it will support native trim/discard, have appropriate overprovisioning, and use internal wear leveling by default. I've done a good bit of Kernel dev for Android. however, since last few years we seriously addressed the problems. EXT4 is better in the general case. 1. XFS is a high-performance, journaling file system designed for high scalability. First of all, some background history. It has lower performance than tried and true ext4 but that is the cost to pay for the features it has. At 16 threads it was a draw (2036 tps vs. Here are some more benchmarks. We recommend btrfs for testing, development, and any non-critical deployments. 10 's new experimental ZFS desktop install option in opting for using ZFS On Linux in place of EXT4 as the root file-system, here are some quick benchmarks looking at the out-of-the-box performance of ZFS/ZoL vs. XFS, like Ext4, is a journaling filesystem. Ext4 limits the number of inodes per group to control fragmentation. 0 SSD for some reference data of the relative F2FS vs. Le système de fichiers ext4 est toujours pris en charge par Red Hat Enterprise Linux 7 et peut être sélectionné au moment de l'installation. AFAIK conclusion 2 is true: ext2/ext3/ext4 are drivers that share a significant part of their code. It presents the. you can chroot, but you won't really have a performance issue with the native WSL drive. Between 2T and 4T on a single disk, any of these would probably have similar performance. Você pode então configurar a aplicação de cotas usando uma opção de montagem. EXT4 run a lot slower when we perform same SQL insert test; XFS respond a lot healthier at 2K INSERT + 2K UPDATE while EXT4 only have 59 for both. 4 was performing the best for RAID0 and RAID10 modes while with RAID1, XFS was performing the best. XFS. Btrfs is one of the most. checksum verification on each file. The ZFS file system combines a volume manager and file. ext4 and also reiserfs store files in a different way. Improve this answer. 0 causes performance drop in ~30-80%. ReiserFS is another filesystem common to linux systems, but with some ongoing codebase issues whereby it periodically tries to kill your wife. Optane SSD RAID Performance With ZFS On Linux, EXT4, XFS, Btrfs, F2FS Storage : 2019-06-20: Linux 5. For more than 3 disks, or a spinning disk with ssd, zfs starts to look very interesting. > I’m a blockquote. A conventional RAID array is a simple abstraction layer that sits between a filesystem and a set of disks. 3. 21 merge window (now known as Linux 5. 61 Comments SSD Disk Observations. Each of the five file-systems were tested on the same NVM Express SSD from the Linux 4. At 16 threads it was a draw (2036 tps vs. It has been suggested that ZFS may not be optimal for fread/fwrite operations, and it may be advisable to utilize ZFS for non-root directories while utilizing ext4 for the remainder of the system for optimal. So for a large video collection, I think I will stick with ext4 still. F2FS vs. XFS is better in general with WT, as the MongoDB production notes suggest. For a future article will be a look at non-mainlined file-systems, including ZFS On Linux. 10 's new experimental ZFS desktop install option in opting for using ZFS On Linux in place of EXT4 as the root file-system, here are some quick benchmarks looking at the out-of-the-box performance of ZFS/ZoL vs. 0 File-System Benchmarks: Btrfs vs. Using Btrfs, just expanding a zip file and trying to immediately enter that new expanded folder in Nautilus, I am presented with a “busy” spinning graphic as Nautilus is preparing to display the new folder contents. Btrfs is one of the most popular newly created file systems, and was. Ceph's recommendation for the choice of filesystem is between btrfs and XFS. Packs several small files into same blocks, conserving filesystem space. The file-systems being benchmarked here are EXT4, XFS, and Btrfs. Primitives for freezing and unfreezing the filesystem for dumping. In our experience Kafka is known to have index failures on such file systems. – in the case of NVMe and regular ext4 with kernel 5. Defaults: ext4 and XFS. Let’s go through the different features of the two filesystems. 77. ext4 파일 시스템은 Red Hat Enterprise Linux 5에서 사용 가능한 기본 ext3 파일 시스템의 확장된 버전입니다. To be honest I'm a little surprised how well Ext4 compared with exFAT ^_^. If we apply a fix by mounting ext4 with dioread_nolock or use xfs, throughput looks good. ) – improvements, bugfixes. 1-based Bcachefs-dev kernel. 79 1. IOSTAT also showing EXT4 was at 98. #filesystem #ext4 #xfs #linuxExplicación de las diferencias entre sistemas de archivos, en este vídeo se comparan los 2 mas usados en GNU/Linux. 8. The test data shown in the graphs below show modest differences between both. 0. 8 snapshot as of last week. 2, and 4. ZFS can vary depending on your specific use case. If you dig in to its history, you will see SGI was famous for workstations designed for audio and video editing. When a copy-on-write is needed, the driver searches through the image's layers to find the right file, starting from the topmost layer. 3. In general, Ext3 or Ext4 is better if an application uses a single read/write thread and small files, while XFS shines when an application uses multiple read/write threads and bigger files The question is XFS vs EXT4.