Projects / phalanx

phalanx

phalanx computes a digest of many buffers simultaneously, and produces a combined hash of them all. It is an initiative to provide a fast, simple, and portable alternative method to compute a checksum in a parallel fashion. It has options for I/O buffer size, hash width, number of threads, and more. It can be run single-threadedly for performance comparisons. It can check files against previously-saved sums, like "MD5sum" does. It also has a "demo" mode, to ascertain accurate operation. It is intended to be useful on large files and multicore/multiprocessor/multithreaded environments.

Tags
Licenses
Operating Systems
Implementation
Translations

Recent releases

  •  27 Jul 2012 01:20

    Release Notes: This release brings many improvements: a "less strong" (but faster) hash method, a "benchmark"-like report in CSV format with optional header, a fix for global log file command-line argument, improved CPU detection on Linux and Cygwin, and even bigger (~1GiB) 'demo' tests.

    •  24 Jul 2012 00:31

      Release Notes: This version implements hash width recognition using the executable file name (e.g. 'pha256sum'), and has an updated man(1) page.

      •  21 Jul 2012 22:39

        Release Notes: This release implements hashing from standard input and adds a proper man(1) page.

        •  19 Jul 2012 21:47

          Release Notes: Starting with this version, PHA is ready for production on all tested platforms and operating systems. This version successfully implements the spice-pool for the stronger hash method, and has a fix for the IA64 port.

          •  19 Jul 2012 00:47

            Release Notes: The '--check' option was fixed on x86.

            Screenshot

            Project Spotlight

            OpenStack4j

            A Fluent OpenStack client API for Java.

            Screenshot

            Project Spotlight

            TurnKey TWiki Appliance

            A TWiki appliance that is easy to use and lightweight.