DragonFly BSD 6.2.2

Published by

DragonFly BSD belongs to the same operating system class as other BSD-derived systems and Linux. It is based on UNIX ideals and APIs and shares ancestor code with other BSD operating systems.
DragonFly BSD belongs to the same operating system class as other BSD-derived systems and Linux. It is based on UNIX ideals and APIs and shares ancestor code with other BSD operating systems.

DragonFly BSD provides an opportunity for the BSD base to grow in an entirely different direction from the one taken in the FreeBSD, NetBSD, and OpenBSD series. DragonFly includes many useful features that differentiate it from other operating systems in the same class.

The most prominent one is HAMMER, our modern high-performance filesystem with built-in mirroring and historic access functionality.

Virtual kernels provide the ability to run a full-blown kernel as a user process for managing resources or for accelerated kernel development and debugging.

The kernel uses several synchronizations and locking mechanisms for SMP. Since the project began, much of the work done has been in this area. A combination of intentional simplification of certain classes of locks to make more expansive subsystems less prone to deadlocks, and the rewriting of nearly all the original codebase using algorithms designed specifically with SMP in mind, has resulted in an extremely stable, high-performance kernel that is capable of efficiently using all CPU, memory, and I/O resources thrown at it.

DragonFly BSD has virtually no bottlenecks or locks contention in-kernel. Nearly all operations can run concurrently on any number of CPUs. Over the years, the VFS support infrastructure (name cache, vnode cache), user support infrastructure (uid, gid, process groups, sessions), process and threading infrastructure, storage subsystems, networking, user and kernel memory allocation, and management, process fork, exec, and exit/teardown, timekeeping, and all other aspects of kernel design have been rewritten with extreme SMP performance as a goal.

DragonFly BSD is uniquely positioned to take advantage of the wide availability of affordable Solid Storage Devices (SSDs) by using swap space to cache filesystem data and meta-data. This feature, commonly called "swap cache," can significantly boost server and workstation workloads with a minor hardware investment.

The DragonFly storage stack comprises robust, natively written AHCI and NVME drivers, stable device names via DEVFS, and a partial implementation of Device Mapper for reliable volume management and encryption.

Some other features that are especially useful to system administrators are a performant and scalable TMPFS implementation, an extremely efficient NULLFS that requires no internal replication of directory or file vnodes, our natively written DNTPD (NTP client), which uses full-bore line intercept and standard deviation summation for highly-accurate timekeeping, and DMA, designed to provide low-overhead email services for system operators who do not need more expansive mail services such as postfix or Sendmail.


  Download