journaled file system: SGI XFS, Reiserfs, IBM JFS, EXT3FS

log files are files that contain system messages, including kernels, services, applications running on the system, and so on. Different log files record different information. The log file system is more secure than the traditional file system because it tracks changes in disk content with separate log files. For example, there are default system log files, some are for security messages only, and some log for cron tasks. Log files can be useful when you are trying to diagnose and troubleshoot system problems, such as attempting to load kernel drivers or looking for unauthorized attempts to use the system. This section discusses where to look for log files, how to view log files, and what to look at in the log file. Some log files are controlled by the daemon called SYSLOGD. A list of log messages maintained by SYSLOGD can be/etc/found in the syslog.conf configuration file. Although Linux can support a wide variety of file systems, almost all modern Linux distributions use EXT4 as the default file system. Ext4 's designers are primarily concerned with file system performance issues. Ext4 writes the contents of a file without simultaneously writing a meta-data (Information about the file, such as permissions, owner, and time of creation and access). In other words, Linux writes the contents of the file first, and then waits until the time is available to write the file's meta-data. This occurs when the contents of the file are written, but the meta-data before the system suddenly loses power, it can cause the file system will be in an inconsistent state. This can lead to serious consequences in a system with a large number of file operations. The log file system is more secure than the traditional file system because it tracks changes in disk content with separate log files. Just like a relational database (RDBMS), the log file system can commit or revoke changes to the file system in a transactional manner. The log file system has important requirements in enterprise-class servers that emphasize data integrity and is the direction of file system development. The idea of a journaled file system comes from large databases such as Oracle. Database operations are often made up of multiple dependent, dependent sub-operations, and any failure of a child operation means that the entire operation is invalid, and any modification to the database data will revert to the previous state of the operation. The journaled file system uses a similar technique. In the partition has a log record file, the file system write operation is the first to operate the record file, if the entire write operation for some reason (such as the system power down) and interrupted, then the next time the system starts to read the contents of the log record file to restore the unfinished write operation, this process generally only takes a few seconds. SGI's XFS file system SGI recently released XFS as1.0. Version 1. SGI's XFS1.0 is available for Linux 2.4. x core Support. It works well on Redhat, and SGI has developed an installer on Redhat. Other Linux distributions are also being developed. SGI's XFS is known as the industry's most advanced, most scalable file system technology. It is a full 64-bit, fast and robust log file system for many years for SGI's IRIX operating system. SGI decided to support the Linux community by empowering Linux with critical infrastructure technologies. It publishes the source code of the XFS they own, and begins porting it in the form of open resources. The work is progressing quickly and is now in the beta phase. As a 64-bit filesystem, XFS can support very large numbers of files (9GX1GB, even larger 18GX1GB), providing significant performance in large 2d and 3d data. XFS has the ability to predict other file system weaknesses while XFS provides enhanced reliability and rapid incident recovery without impeding performance. SGI's XFS offers new features for the Linux and open-source communities: Upgradeable XFS is designed to be upgradeable to face most storage capacity and I/o storage requirements that can handle large files and large directories containing huge amounts of files to meet the rapidly growing disk requirements of the 21st century. XFS has the ability to dynamically allocate index space to files, enabling the system to form the ability to efficiently support large numbers of files. With its support, users can use 1exabyte (1GX1GB) large files, much larger than the current largest file system. Excellent I/o Performance typical modern servers use large striped disk arrays to provide up to several gigabytesThe total bandwidth per second. XFS can well satisfy the size and concurrency of I/O requests i/o the number of requests. System troubleshooting XFS can recover from most unexpected outages in a second. Traditional file systems must perform specific file system checks after an outage, which can take several hours to complete. XFS avoids lengthy file system checks that can significantly reduce the time it takes to read and write to the disk. XFS is available as the root filesystem and is supported by Lilo. It's okay to use on NFS servers. The home page of the ReiserFS file system that supports software RAID and Lvm.reiserfs file systems is on July 23, 1997, Hans Reiser his ReiserFS file system based on the balanced tree structure on the web. This is ReiserFS's first public appearance. Since then, ReiserFS has been developing and developing under Hans Reiser and under the leadership of the development team, and SuSE Linux has also contributed significantly to its development. Since ReiserFS has some very useful features, the main one is that it is much faster than EXT2FS, so it is quickly used by many people. Linux kernel 2.4.1-The Pre4 version starts with the ReiserFS file system. The current reiserfs can only be used on Intel architectures, but ReiserFS versions that support other architectures are also being actively developed. The new generation of Reiser4 will be in 2002/9/released 30th. ReiserFS is a new type of file system that you can choose to load your Linux operating system and other applications and data files in blue Dot Linux. It's a different way of doing it.--fully balanced tree structure to accommodate data, including file data, filenames, and log support. ReiserFS also supports massive disk and disk arrays, and can continue to maintain fast search speeds and high efficiency on top. ReiserFS is a file system structure based on the balance tree, especially for large files of huge file system, such as the file system on the server, the search speed is faster than Ext2, Ext2 uses the local binary search method, the comprehensive performance is inferior to reiserfs. Space allocation and utilization The directory in ReiserFS is fully dynamically allocated, so there is no case where the disk space that is common in ext2 cannot be reclaimed by a huge directory. Small files in ReiserFS (<4K) can be stored directly into the tree, small file read and write faster, the tree nodes are byte-aligned, small files can share the same hard disk block, save a lot of space. EXT2 uses a fixed-size block allocation strategy, which means that less than 4K of small files also occupy 4K of space, resulting in a more serious space waste. Advanced logging mechanism ReiserFS has advanced logging (journaling/logging) Mechanism, incomplete file operations do not affect the integrity of the entire file system structure when the system crashes unexpectedly. Ext2 is robust, but once the file system is disconnected, it will have to perform a lengthy process of checking the integrity of the system's data structures at the next boot, which is necessary to prevent data loss. For larger server file systems, this type of"File System Check"It may take several hours, and in many cases such a long time is unacceptable. A technique to solve this problem"Log File System". With the help of the log, each change to the data structure is recorded, and the log is guaranteed to be written to the hard disk before each actual data modification. Because of this, when the system suddenly crashes, a few seconds after the next boot can be restored to a complete system, the system can be used quickly. Support for massive disks and excellent integrated performance ReiserFS is a fairly modern file system, in contrast, ext2, although it has been very good, but its design is only the 1880s standard. The advent of ReiserFS, so that Linux has a like Irix/aix high-grade commercial UNIX has the advanced file system. ReiserFS can easily manage hundreds of file systems in the enterprise application, because of its efficient storage and fast small file I/O characteristics, it is also outstanding on the desktop system: the time to start the X-Window system ReiserFS less than ext2 1/3. Ext2 is not able to manage a single file above 2G, which makes ReiserFS better than ext2 in some enterprise-class applications. ReiserFS one of the most criticized shortcomings is that each version of the upgrade will have to reformat the disk once, and this drawback is being improved. IBM's JFS file system IBM latest release XFS1.0. Version 2. JFS (journaled File System Technology forLinux) Developers include the major developers of the JFS of Aix (IBM Unix). On AIX, JFS has withstood the test. It is reliable, fast and easy to use. In February 2000, IBM announced the porting of the Linux version of the JFS file system under an open resource license. JFS is also an enterprise-class file system with a large number of user installations. With the log file system, there is a slight performance penalty because the system needs to write a lot of data when it keeps a log. But it's worth it to crash the system and run fsck. JFS reduces the loss of performance by converting the synchronization log to an asynchronous log. With synchronization logs, the system directly makes logging operations proportional to the action of the file system, while asynchronous logs, which have fewer log records and can be logged without affecting other file system operations, can reduce the cost of logging. Iv. ext2 successor Ext3fsext3 is a log file system developed by the open resources community, the main developer is Redhat hacker master Stephen Tweedie. Ext3 is designed to be an upgraded version of Ext2, allowing users to migrate from EXT2FS to Ext3fs as easily as possible. Ext3 the log function of record metadata on the basis of ext2, and strives to keep forward and backward compatibility. This file system may be called the next version of Ext2 more appropriately. Ext3fs work very well. The biggest advantage of EXT3FS is backwards compatible with ext2. And EXT3FS also supports asynchronous logging, which means it might perform better than ext2. However, when using the Ext3 file system, you should pay attention to the hard disk quota problem, before this problem is resolved, it is not recommended to use EXT3 in important applications .+disk QUOTAEXT3 is the latest version for kernel2.4.8 of ext3-2.4-0.9.6-248. GZ's newly launched Redhat7The Ext3,turbolinux Server is officially enabled on the. 2B6. 5 supports both ReiserFS and Ext3. By adding a log file to an existing Ext2 file system, the Ext2 file system can be converted to a ext3 file system. You could switch the file system to ext3 and ext2 freely. Because the ext2 log file on the Ext3 file system still exists, Just ext2 can't recognize it.

journaled file system: SGI XFS, Reiserfs, IBM JFS, EXT3FS