All of lore.kernel.org
 help / color / mirror / Atom feed
* question on fio memory handling
@ 2019-01-08 10:44 Paolo Valente
  0 siblings, 0 replies; only message in thread
From: Paolo Valente @ 2019-01-08 10:44 UTC (permalink / raw)
  To: fio; +Cc: Alessio Masola

Hi,
we are using fio to generate interleaved I/O, by making several job read from the whole /dev/sda and we are using zoneskip to create the interleave among jobs.

But we have bumped into the following issue, shown for just one job. The following command:

$ sudo fio --name=reader0 -rw=read --filename=/dev/sda --zonesize=16384 --zonemode=zbd --zoneskip=16384
smalloc: OOM. Consider using --alloc-size to increase the shared memory available.

In contrast, if we invoke,

sudo fio --name=reader0 -rw=read --filename=/dev/sda --zonesize=16384 --zonemode=strided --zoneskip=16384

(which is ok for our goal at the moment), then we everything seems to work fine.

Could you please example me why --zonemode=zbd fails by apparently trying to allocate memory for the whole size of the special file /dev/sda? Is it (strictly) necessary for doing random I/O among zones?

Thanks,
Paolo

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2019-01-08 10:44 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-01-08 10:44 question on fio memory handling Paolo Valente

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.