Journaling with ReiserFS
Mail servers tend to be a worst-case for the journaled file systems because they need to make sure each file operation completes. They use fsync, or a bunch of other tricks to prevent losing messages after a crash, and this forces the file system to close transactions while they are still very small.
Mail servers need a fast way to get new files committed to disk, and data logging can help with this. During the fsync call, you log the data blocks and metadata required to add the file in the tree, producing one large sequential write. If the file being written is a transient spool file, it might never be written back to the main drive. Combined with a fast dedicated logging device, data logging can make a big performance improvement.
As of yet, the ReiserFS 2.4 code does not support data logging. There is data logging code in our releases for the 2.2 kernels, but it needs to be adapted to the 2.4 page cache.
The end result of the new Linux file systems should be very interesting. Admins will be able to choose the best product for their applications, and programmers will be able to compare their decisions against the alternatives. Linux as a whole will benefit as the community picks and chooses the best features from each file system.
Chris Mason (email@example.com) was a system administrator before he started contributing to the ReiserFS Project. He now works full-time for SuSE from his home in Rochester New York.
|Non-Linux FOSS: libnotify, OS X Style||Jun 18, 2013|
|Containers—Not Virtual Machines—Are the Future Cloud||Jun 17, 2013|
|Lock-Free Multi-Producer Multi-Consumer Queue on Ring Buffer||Jun 12, 2013|
|Weechat, Irssi's Little Brother||Jun 11, 2013|
|One Tail Just Isn't Enough||Jun 07, 2013|
|Introduction to MapReduce with Hadoop on Linux||Jun 05, 2013|
- Containers—Not Virtual Machines—Are the Future Cloud
- Non-Linux FOSS: libnotify, OS X Style
- Lock-Free Multi-Producer Multi-Consumer Queue on Ring Buffer
- Linux Systems Administrator
- Introduction to MapReduce with Hadoop on Linux
- RSS Feeds
- New Products
- Validate an E-Mail Address with PHP, the Right Way
- Weechat, Irssi's Little Brother
- Tech Tip: Really Simple HTTP Server with Python
- Poul-Henning Kamp: welcome to
20 min 43 sec ago
- This has already been done
21 min 43 sec ago
- Reply to comment | Linux Journal
1 hour 6 min ago
- Welcome to 1998
1 hour 55 min ago
- notifier shortcomings
2 hours 19 min ago
3 hours 56 min ago
- Android User
3 hours 57 min ago
- Reply to comment | Linux Journal
5 hours 50 min ago
8 hours 40 min ago
- This is a good post. This
13 hours 53 min ago
Free Webinar: Hadoop
How to Build an Optimal Hadoop Cluster to Store and Maintain Unlimited Amounts of Data Using Microservers
Realizing the promise of Apache® Hadoop® requires the effective deployment of compute, memory, storage and networking to achieve optimal results. With its flexibility and multitude of options, it is easy to over or under provision the server infrastructure, resulting in poor performance and high TCO. Join us for an in depth, technical discussion with industry experts from leading Hadoop and server companies who will provide insights into the key considerations for designing and deploying an optimal Hadoop cluster.
Some of key questions to be discussed are:
- What is the “typical” Hadoop cluster and what should be installed on the different machine types?
- Why should you consider the typical workload patterns when making your hardware decisions?
- Are all microservers created equal for Hadoop deployments?
- How do I plan for expansion if I require more compute, memory, storage or networking?