"HAMMER makes no modifications to the B-Tree whatsoever on the front-end. When you create, delete, rename, write, etc... when you do those operations HAMMER caches them in a virtualization layer in memory and doesn't make any modifications to its on-media data structures (or their in-memory representations) at all until the meta-data is synced to disk," DragonFly BSD creator Matthew Dillon explained, comparing HAMMER, his clustering filesystem, to a wiki summary of Reiser4's implementations. He continued:

"HAMMER uses a modified B+Tree for its on-disk representation, which is a B-Tree with only keys at internal nodes and only records at the leafs. This was done to reduce structural bloat, allow for a leaf->leaf linking optimization in the future, and for other reasons. [...] HAMMER's internal nodes have a left and right bounding element. A standard B+Tree only has a left bounding element. By adding a right bounding element HAMMER can cache pointers into its B+Tree and 'pick up' searches, insertions, and deletions relative to the cached pointers instead of having to start at the root of the tree. More importantly, it can pickup searches, insertions, and deletions at internal nodes, not just leaf nodes. So I can cache a proximity pointer and if I do a good job I never have to traverse the B+Tree above that point."

"I regularly run and post various benchmarks comparing POHMELFS, NFS, XFS and Ext4, [the] main goal of POHMELFS at this stage is to be essentially as fast as [the] underlying local filesystem. And it is..." explained Evgeniy Polyakov, suggesting that the POHMELFS networking filesystem performs 10% to 300% faster than NFS, depending on the file operation. In particular, he noted that it still suffers from random reads, an area that he's currently focused on fixing. He summarized the new features found in the latest release:

"Read request (data read, directory listing, lookup requests) balancing between multiple servers; write requests are sent to multiple servers and completed only when all of them send an ack; [the] ability to add and/or remove servers from [the] working set at run-time from userspace; documentation (overall view and protocol commands); rename command; several new mount options to control client behaviour instead of hard coded numbers."

Looking forward, Evgeniy noted that this was likely the last non-bugfix release of the kernel client side implementation, suggesting that the next release would focus on adding server side features, "needed for distributed parallel data processing (like the ability to add new servers via network commands from another server), so most of the work will be devoted to server code."

"Oh great, not yet-another-kernel-tree, just what the world needs..." began Greg KH, continuing, "yes, this is an announcement of a new kernel tree, linux-staging." He explained:

"In a long and meandering thread with some of the other kernel developers a week or so ago, it came up that there is no single place for companies and developers to put their code for testing while it gets cleaned up for submission into the kernel tree. All of the different subsystems have trees, but they generally only want code that is about to go into this release, or the next one. For stuff that is farther off, there is no place to go. So, here's the tree for it."

In a readme created for the new tree, Greg adds, "the linux-staging tree was created to hold drivers and filesystems and other semi-major additions to the Linux kernel that are not ready to be merged at this point in time." He also requested that the new tree be included in Linux -next, leading Theodore Ts'o to ask, "does this mean that the nature of linux-next is changing? I thought the whole point of linux-next was only to have what would be pushed to Linus in the near future, so we could check for patch compatibility issues." Greg explained that he was hoping for an exception for his new -staging tree as it only includes whole new drivers and filesystems, not changes to existng features, "there is stuff that users can use to get hardware to work that currently is not supported on kernel.org kernels at all." As an example he noted, "there are 2 big network drivers in there that support a wide range of devices that some people would like to see working :)"

"I'd like to say that the diffs are shrinking and things are calming down, but I'd be lying," began Linux creator Linus Torvalds, announcing the 2.6.26-rc6 kernel. He noted, "another week, another -rc, and another 350 commits. Yes, the diff is smaller than the one from rc4 to rc5 (despite having more commits), so we're on the right trajectory, but I was hoping for less churn at this stage." Linus continued:

"As usual, most of the changes are to drivers (with arch updates a strong second). The DVB updates are the biggest chunk of that, but on the whole it's quite spread out. As mentioned, the diffs are smaller and there are more commits, and yes, most of the commits are really rather small and trivial fixes.

"Give it a try, we should have a few less regressions once more,"

"After another round of performance tuning HAMMER all my benchmarks show HAMMER within 10% of UFS's performance, and it beats the shit out of UFS in certain tests such as file creation and random write performance," noted DragonFly BSD creator Matthew Dillon, providing an update on his new clustering filesystem. He continued, "read performance is good but drops more then UFS under heavy write loads (but write performance is much better at the same time)." He then referred to the blogbench benchmark noting, "now when UFS gets past blog #300 and blows out the system caches, UFS's write performance goes completely to hell but it is able to maintain good read performance." Matthew then compared this to HAMMER:

"HAMMER is the opposite. It can maintain fairly good write performance long after the system caches have been blown out, but read performance drops to about the same as its write performance (remember, this is blogbench doing reads from random files). Here HAMMER's read performance drops significantly but it is able to maintain write performance. UFS's write performance basically comes to a dead halt. However, HAMMER's performance numbers become 'unstable' once the system caches are blown out."

"These patches allow data integrity information (checksum and more) to be attached to I/Os at the block/filesystem layers and transferred through the entire I/O stack all the way to the physical storage device," began Martin Petersen. He went on to explain, "the integrity metadata can be generated in close proximity to the original data. Capable host adapters, RAID arrays and physical disks can verify the data integrity and abort I/Os in case of a mismatch." He noted that support currently only exists for SCSI disks, but that work is underway to also add support for SATA drives and SCSI tapes, "with a few minor nits due to protocol limitations the proposed SATA format is identical to the SCSI". Explaining how this works, Martin continued:

"SCSI drives can usually be reformatted to 520-byte sectors, yielding 8 extra bytes per sector. These 8 bytes have traditionally been used by RAID controllers to store internal protection information. DIF (Data Integrity Field) is an extension to the SCSI Block Commands that standardizes the format of the 8 extra bytes and defines ways to interact with the contents at the protocol level. [...] When writing, the HBA (Host Bus Adapter) will DMA 512-byte sectors from host memory, generate the matching integrity metadata and send out 520-byte sectors on the wire. The disk will verify the integrity of the data before committing it to stable storage. When reading, the drive will send 520-byte sectors to the HBA. The HBA will verify the data integrity and DMA 512-byte sectors to host memory."

In a series of seven patches, Arnd Bergmann proposed adding in-memory write support to mounted cramfs file systems. He explained, "the intention is to use it for instance on read-only root file systems like CD-ROM, or on compressed initrd images. In either case, no data is written back to the medium, but remains in the page/inode/dentry cache, like ramfs does." Reactions were mixed. When Arnd suggested this as an alternative to using the more complex unionfs to overlay a temporary filesystem over a read-only file system, and that similar support could be added to other file systems, it was pointed out that there was ultimately more gained by focusing on a single solution that worked with all filesystems. David Newall stressed, "multiple implementations is a recipe for bugs and feature mismatch." Erez Zadok suggested, "I favor a more generic approach, one that will work with the vast majority of file systems that people use w/ unioning, preferably all of them." He went on to add that more gains would be had from modifying the union destination filesystem rather than multiple source filesystems. Arnd agreed in principle, but noted it would add complexity. He indicated that he'd explore the idea further, then explained:

"My idea was to have it in cramfs, squashfs and iso9660 at most, I agree that doing it in even a single writable file system would add far too much complexity. I did not mean to start a fundamental discussion about how to do it the right way, just noticed that there are half a dozen implementations that have been around for years without getting close to inclusion in the mainline kernel, while a much simpler approach gives you sane semantics for a subset of users."