During perf test we figured out, that global qd
settings is split between disks number in filename.
Setting this parameter (multiplied by section disks number
and numjobs) in filename section give as expected value.
Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: I89a44e48f10da131e2e9128f3ff2fbfb197076ad
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2749
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Skip copying and building SPDK on host systems if
specified in configuration file.
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: I4f9c0dd77e020dfd866f52dc138526e63dd9222d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3075
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Aggregate separate read/write results into
a single stat. Reduces the number of additional
fields and formulas in spreadsheets when preparing
performance reports.
Change-Id: I5fdab35c1bb69aad98f7b9808885801db0416449
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3068
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Reviewed-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Test result IOPS and BW are a sum from across all initiators,
but latencies should be an average from accross initiators.
In previous version it was mistakenly a sum, and that had
to be manually handled in Excel spreadsheets when gathering
data for performance reports.
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: I17e38e3cdd593dcda9730c19decc2d6a72a11130
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3067
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Save each initiator results to a .csv file. Useful
for a bit more insight than aggregated results.
Change-Id: I1fafe52ba78f2c62abfd0e3c863505515a363cd3
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3053
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Maciej Wawryk <maciejx.wawryk@intel.com>
I suggest putting "make clean" before "./configure". Like Migration-tc3a.sh (spdk/test/vhost/migration):
ssh_remote $MGMT_INITIATOR_IP "cd $spdk_repo_share_dir/spdk; make clean; ./configure --with-rdma --enable-debug; make -j40"
Otherwise, once the configuration is changed, some outdated compiled files may remain after "make clean".
Change-Id: I7300749986129cbfaef44f8ff72fad0449f4e081
Signed-off-by: WANGHAILIANG <hailiangx.e.wang@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2805
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: <dongx.yi@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Liang Yan <liang.z.yan@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
This makes them easier to find. They are now named
spdk_nvme and spdk_bdev.
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Change-Id: I1f6736d8f7f5b9669583731fb0b710ba54f0d50a
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2679
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Automatically place binaries produced from the app directory
into build/bin. This matches with the output in build/lib
that already exists.
Change-Id: I13cd2da71d2f88592e22308fe8a907bf458458b5
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2379
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
With issues similar to https://github.com/spdk/spdk/issues/1293 it
would be better to use temporary directories for spdk workspace
instead of depending on the hardcoded path (/tmp/spdk). This will
mitigate potential scenario where permissions of the target dir
are suddenly changed in-between autobuild runs.
Change-Id: If44a2dcce712a185287186f33e7361223dc6d451
Signed-off-by: Michal Berger <michalx.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/1863
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: Icf45fbdff547fcf45c361b1d9c95b352756ad126
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/1775
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This will allow us to use num_cores and cpus_allowed option also
in Kernel initiator side.
Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: Iad4e8bf7cd8a614ada8f443a8b1036dc2c4803e4
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/1173
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
PCM cpu measure give as result 1306 columns, we want check only ones which contains
SKTdata and SKTtraffic values.
Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: I52b6781a7ff99e0d894f7f1c2b25b655fa274eb1
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/946
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
PCM - Processor Counter Monitor is an application to monitor performance
and energy metrics of Intel® Core™, Xeon®, Atom™ and Xeon Phi™ processors.
Using this measure tool will help us to capture memory bandwidth consumed
on Intel processors.
Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: Ie580d2aad6ddd98c5256df65f65e478dd78317e5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/541
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Removing hardcoded "-i 8" value for kernel NVMe-oF initiator
parameter "nr-io-queues".
This was previously merged by mistake and had to be removed
manually in test runs.
Change-Id: I454ad468863f93ac2ce93a1f0bd00f678ed4995c
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/477766
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Reviewed-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Reviewed-by: Seth Howell <seth.howell@intel.com>
Move code around so that lines containing arguments
for constructors are better aligned with how they're
inherited.
Change nvmet_dir and fio_dir to nvmet_bin and fio_bin to
point directly to executable files. There is no advantage
in using directories.
Additionally fix how fio_bin (previously fio_dir) is passed
to parent classes - it was broken and caused exceptions.
Update README.md accordingly.
Change-Id: I10524593d543e236bf318a8cd51ea5375c334e8b
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/477015
Community-CI: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Reviewed-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Added Initiator configuration option to point to the
fio source directory used when compiling SPDK with the --with-fio
option and run SPDK test with the fio plugin.
Signed-off-by: John Kariuki <John.K.Kariuki@intel.com>
Change-Id: I097cb17691a72d7c2a0e198dccd12cf3d1c3474c
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/468890
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Fixed the following:
1) Missing python paramiko module in package dependecies.
2) Paramiko connection error: TypeError: missing_host_key() missing 1 required positional argument: 'key'
The set_missing_host_key_policy call should take a class instance AutoAddPolicy(), instead of the class name.
3) JSON config file defines "rdma_ips" instead of "nic_ips"
4) Added result file location in the documentation
5) Added env variable for PYTHONPATH to documentation
6) config.json changed default rw to randrw instead of read
Change-Id: I96624e2912131f62254c684a6c03a53a7f806cde
Signed-off-by: John Kariuki <John.K.Kariuki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/455285
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Liang Yan <liang.z.yan@intel.com>
For future NVMeOF benchmarking scripts must support TCP layer.
This is a recent feature so also added support of using custom
built nvme-cli.
Example config file was updated with appropiate fields.
Rename rdma_ips to nic_ips, as NICs no longer need to be
RDMA-enabled for this test.
Change-Id: I368b48eca08a30b4221f198cb76134b8836dbba9
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/449879
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Wrong indent level caused results to be duplicated in final
output file if configuration file was used to run more than one
workload combination.
Change-Id: I0fadc10eba38898df648ec2da5fbd0440a3b9693
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/449150
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
