Provided by: blkreplay_1.0-3build1_amd64 bug


       blkreplay - block device testing and benchmarking tool


       blkreplay [options] device


       blkreplay is a utility driving the block layer of the operating system while measuring
       latency and throughput of I/O operations for later visualisation.

       blkreplay can create artificial loads (random read-write sweeps, various kinds of overload
       tests) or replay natural loads which have been recorded by blktrace or a similar utility
       run at production servers.

       blkreplay can be used to test physical hardware, to compare different brands of hard disks
       or RAID controllers, to evaluate the effect of SSD caching, to compare different block
       level transports like iSCSI vs Fibrechannel and so on.

       blkreplay parses the load data from stdin. To create load data from blktrace output, use script.


       Influence replay duration:
               start offset (in seconds, 0=from_start)

               end offset (in seconds, 0=unlimited)

               alternatively specify the end offset as delta

               start offset, used for output (in seconds)

               start after grace period for filling the pipes (in seconds)

       Handling of conflicting I/O requests:
               conflicting writes are ALLOWED (damaged I/O)

               conflicting writes are simply dropped

               partial ordering by pushing back conflicts (default)

               enforce total order in case of conflicts

               mode between 0 and 2, see docs (default=1)

       Replay parameters:
               parallelism (default=1024)

               fill data blocks with random bytes (%, default=0)

       Verification modes:
               verify is OFF (default)

               verify on reads

               additional verify pass at the end

               re-read after each write (destroys performance)

               increase verbosity, show additional INFO: output

       Expert options (DANGEROUS):
               use O_DIRECT (default)

               don't use O_DIRECT, deliver FAKE results

               use O_SYNC

               don't use O_SYNC (default)

               don't actually do I/O, measure internal overhead

               omit lseek() and tags, even less internal overhead

               delay value for I/O simulation (timespec sec.nsec)

               limit pipe fillahead (realtime sec.nsec)

               only for kernel hackers (default=4)

               only for kernel hackers

               max #requests on dispatch

               speedup / slowdown by REAL factor (default=1.0)

               use mmap() instead of read() / write() [NYI]


       blkreplay was written by Thomas Schoebel-Theuer. This manual page was created from
       blkreplay documentation by Andrew Shadura.


       Copyright 2009-2012 Thomas Schoebel-Theuer



                                            2018-04-03                               BLKREPLAY(1)