Re: [Dovecot] Best inode_ratio for maildir++ on ext4
On 19/02/2010 23:51, Stan Hoeppner wrote: Rodolfo Gonzalez put forth on 2/19/2010 5:18 PM: Hi, This might be a silly question: which would be the best inode ratio for a 5 Tb filesystem dedicated to Maildir++ storage? I use ubuntu server, which has a preconfigured setting for mkfs.ext4 called news with inode_ratio = 4096, and after formating the fs with that setting and then with the defautl setting I see this difference of space (wasted space, but more inodes): 4328633696 free 1K-blocks with mkfs's -T news switch = 1219493877 free inodes 4557288800 free 1K-blocks with default mkfs settings = 304873461 free inodes I'll be storing e-mail messages for around 20,000 accounts on that partition (average 512 Mb per account). Would you consider worth the waste of about 200 Gb of the filesystem space in exchange of more inodes? If your version of Ubuntu server has XFS support built in, forget ext4 and go XFS. It's more reliable, faster in every single benchmark I've seen especially for large numbers of files, both large and small, has a ton of management tools and instrumentation interfaces, and has a proven enterprise track record. Someone posted about XFS some time back. It apparently can perform well, but it will need some tuning (something to do with number of log buffers from memory?) to perform well on Maildir type operations Search around and benchmark is my suggestion (there are some mail benchmarking programs) Good luck Ed W
Re: [Dovecot] Best inode_ratio for maildir++ on ext4
Rodolfo Gonzalez put forth on 2/19/2010 5:18 PM: Hi, This might be a silly question: which would be the best inode ratio for a 5 Tb filesystem dedicated to Maildir++ storage? I use ubuntu server, which has a preconfigured setting for mkfs.ext4 called news with inode_ratio = 4096, and after formating the fs with that setting and then with the defautl setting I see this difference of space (wasted space, but more inodes): 4328633696 free 1K-blocks with mkfs's -T news switch = 1219493877 free inodes 4557288800 free 1K-blocks with default mkfs settings = 304873461 free inodes I'll be storing e-mail messages for around 20,000 accounts on that partition (average 512 Mb per account). Would you consider worth the waste of about 200 Gb of the filesystem space in exchange of more inodes? If your version of Ubuntu server has XFS support built in, forget ext4 and go XFS. It's more reliable, faster in every single benchmark I've seen especially for large numbers of files, both large and small, has a ton of management tools and instrumentation interfaces, and has a proven enterprise track record. -- Stan
Re: [Dovecot] Best inode_ratio for maildir++ on ext4
Hi! On Fre, 2010-02-19 at 17:18 -0600, Rodolfo Gonzalez wrote: [...] This might be a silly question: which would be Not at all IMHO. the best inode ratio for a 5 Tb filesystem dedicated to Maildir++ storage? I use ubuntu server, which has a preconfigured setting for mkfs.ext4 called news with inode_ratio = 4096, and after formating the fs with that setting and then with the defautl setting I see this difference of space (wasted space, but more inodes): 4328633696 free 1K-blocks with mkfs's -T news switch = 1219493877 free inodes 4557288800 free 1K-blocks with default mkfs settings = 304873461 free inodes I'll be storing e-mail messages for around 20,000 accounts on that partition (average 512 Mb per account). Would you consider worth the waste of about 200 Gb of the filesystem space in exchange of more inodes? That depends entirely if 512MB mail per account a few large ones or a lot of small ones (assuming that the future behaviour is similar to the past). So perhaps it helps to count the files (and directories) on that file system as each of them actually uses an i-node. BTW you can set other values than default and news, namely the number directly. Bernd -- Bernd Petrovitsch Email : be...@petrovitsch.priv.at LUGA : http://www.luga.at
Re: [Dovecot] Best inode_ratio for maildir++ on ext4
On Fri, 2010-02-19 at 17:51 -0600, Stan Hoeppner wrote: If your version of Ubuntu server has XFS support built in, forget ext4 and go XFS. It's more reliable, faster in every single benchmark I've seen especially for large numbers of files, both large and small, has a ton of management tools and instrumentation interfaces, and has a proven enterprise track record. Agree wth XFS, providing, and a big providing, you have reliable and guaranteed power, hard powerouts on XFS are not known for their niceness and protection of data -- Kind Regards, SSA Noel Butler L.C.P No. 251002 This Email, including any attachments, may contain legally privileged information, therefore remains confidential and subject to copyright protected under international law. You may not disseminate or reveal any part to anyone without the authors express written authority to do so. If you are not the intended recipient, please notify the sender and delete all relevance of this message including any attachments, immediately. Confidentiality, copyright, and legal privilege are not waived or lost by reason of the mistaken delivery of this message. Only PDF and ODF documents are accepted, do not send Microsoft proprietary formatted documents.
Re: [Dovecot] Best inode_ratio for maildir++ on ext4
Bugger, hit enter too soon, was going to say, it is probably better than using EXT4 though, why on earth anyone would use that on a serious production server I'll never know. On Sat, 2010-02-20 at 11:51 +1000, Noel Butler wrote: On Fri, 2010-02-19 at 17:51 -0600, Stan Hoeppner wrote: If your version of Ubuntu server has XFS support built in, forget ext4 and go XFS. It's more reliable, faster in every single benchmark I've seen especially for large numbers of files, both large and small, has a ton of management tools and instrumentation interfaces, and has a proven enterprise track record. Agree wth XFS, providing, and a big providing, you have reliable and guaranteed power, hard powerouts on XFS are not known for their niceness and protection of data Kind Regards, SSA Noel Butler L.C.P No. 251002 This Email, including any attachments, may contain legally privileged information, therefore remains confidential and subject to copyright protected under international law. You may not disseminate or reveal any part to anyone without the authors express written authority to do so. If you are not the intended recipient, please notify the sender and delete all relevance of this message including any attachments, immediately. Confidentiality, copyright, and legal privilege are not waived or lost by reason of the mistaken delivery of this message. Only PDF and ODF documents are accepted, do not send Microsoft proprietary formatted documents.
Re: [Dovecot] Best inode_ratio for maildir++ on ext4
Noel Butler wrote: Agree wth XFS, providing, and a big providing, you have reliable and guaranteed power, hard powerouts on XFS are not known for their niceness and protection of data Bugger, hit enter too soon, was going to say, it is probably better than using EXT4 though, why on earth anyone would use that on a serious production server I'll never know. I used to have the maildirs on ReiserFS and never had a problem with it, but given the current state of that FS and that I weren't really comfortable with it, I'll give XFS a try for the maildir array and the postfix queue partition. After formating, I got 4.6 Tb of usable space, which makes me happy, and also the dynamic inode allocation. Regards, Rodolfo. P.S. I have UPS and generator.
Re: [Dovecot] Best inode_ratio for maildir++ on ext4
Rodolfo Gonzalez Gonzalez put forth on 2/20/2010 12:18 AM: I used to have the maildirs on ReiserFS and never had a problem with it, but given the current state of that FS and that I weren't really comfortable with it, I'll give XFS a try for the maildir array and the postfix queue partition. After formating, I got 4.6 Tb of usable space, which makes me happy, and also the dynamic inode allocation. http://en.wikipedia.org/wiki/XFS Like I said, it's a very mature high performance journaled FS with many enterprise level features, dynamic inode allocation being one of many. It was introduced by SGI in 1994 and has been in constant development since then. It was ported to Linux around 2000 and introduced into the mainline kernel in 2.4. It is the only filesystem ever used on SMP servers from 128+ CPUs up to 1024 CPUs. This is because SGI is the only company to ever offer SMP systems beyond 128 CPUs. They are actually ccNUMA, not SMP, but the programming model is SMP, because every CPU in the machine can directly address memory in any NUMA node in the system. The only practical difference between ccNUMA and a true SMP is the memory latency. Obviously, scalability and the ability to manipulate very large filesystems with large numbers of files is required for such massive machines. The Columbia supercomputer at the NASA Ames facility consists of 20 such machines, each with 512 CPUs. The system has a 1 Peta Byte (raw) RAID subsystem formatted with CXFS, the clustered version of XFS. XFS scales very well. ;) I've been fan of SGI for a long time. I could never afford/justify one of their machines. I'm so glad they open sourced XFS and are sharing this fantastic filesystem with the rest of us who could never afford their gear. Many would agree with me if I said it is hands down the overall best *nix filesystem available for most workloads. It's not suitable on Linux for /boot or /, but for just about everything else it is king of the hill. -- Stan
