qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
To: qemu-block@nongnu.org
Cc: qemu-devel@nongnu.org, vsementsov@virtuozzo.com,
	eblake@redhat.com, armbru@redhat.com, fam@euphon.net,
	stefanha@redhat.com, mreitz@redhat.com, kwolf@redhat.com,
	den@virtuozzo.com
Subject: [PATCH v7 14/21] scripts/simplebench: support iops
Date: Wed, 21 Oct 2020 17:58:52 +0300	[thread overview]
Message-ID: <20201021145859.11201-15-vsementsov@virtuozzo.com> (raw)
In-Reply-To: <20201021145859.11201-1-vsementsov@virtuozzo.com>

Support benchmarks returning not seconds but iops. We'll use it for
further new test.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
---
 scripts/simplebench/simplebench.py | 38 ++++++++++++++++++++++--------
 1 file changed, 28 insertions(+), 10 deletions(-)

diff --git a/scripts/simplebench/simplebench.py b/scripts/simplebench/simplebench.py
index 2445932fc2..2251cd34ea 100644
--- a/scripts/simplebench/simplebench.py
+++ b/scripts/simplebench/simplebench.py
@@ -24,9 +24,12 @@ def bench_one(test_func, test_env, test_case, count=5, initial_run=True):
 
     test_func   -- benchmarking function with prototype
                    test_func(env, case), which takes test_env and test_case
-                   arguments and returns {'seconds': int} (which is benchmark
-                   result) on success and {'error': str} on error. Returned
-                   dict may contain any other additional fields.
+                   arguments and on success returns dict with 'seconds' or
+                   'iops' (or both) fields, specifying the benchmark result.
+                   If both 'iops' and 'seconds' provided, the 'iops' is
+                   considered the main, and 'seconds' is just an additional
+                   info. On failure test_func should return {'error': str}.
+                   Returned dict may contain any other additional fields.
     test_env    -- test environment - opaque first argument for test_func
     test_case   -- test case - opaque second argument for test_func
     count       -- how many times to call test_func, to calculate average
@@ -34,8 +37,9 @@ def bench_one(test_func, test_env, test_case, count=5, initial_run=True):
 
     Returns dict with the following fields:
         'runs':     list of test_func results
-        'average':  average seconds per run (exists only if at least one run
-                    succeeded)
+        'dimension': dimension of results, may be 'seconds' or 'iops'
+        'average':  average value (iops or seconds) per run (exists only if at
+                    least one run succeeded)
         'delta':    maximum delta between test_func result and the average
                     (exists only if at least one run succeeded)
         'n-failed': number of failed runs (exists only if at least one run
@@ -54,11 +58,19 @@ def bench_one(test_func, test_env, test_case, count=5, initial_run=True):
 
     result = {'runs': runs}
 
-    succeeded = [r for r in runs if ('seconds' in r)]
+    succeeded = [r for r in runs if ('seconds' in r or 'iops' in r)]
     if succeeded:
-        avg = sum(r['seconds'] for r in succeeded) / len(succeeded)
+        if 'iops' in succeeded[0]:
+            assert all('iops' in r for r in succeeded)
+            dim = 'iops'
+        else:
+            assert all('seconds' in r for r in succeeded)
+            assert all('iops' not in r for r in succeeded)
+            dim = 'seconds'
+        avg = sum(r[dim] for r in succeeded) / len(succeeded)
+        result['dimension'] = dim
         result['average'] = avg
-        result['delta'] = max(abs(r['seconds'] - avg) for r in succeeded)
+        result['delta'] = max(abs(r[dim] - avg) for r in succeeded)
 
     if len(succeeded) < count:
         result['n-failed'] = count - len(succeeded)
@@ -118,11 +130,17 @@ def ascii(results):
     """Return ASCII representation of bench() returned dict."""
     from tabulate import tabulate
 
+    dim = None
     tab = [[""] + [c['id'] for c in results['envs']]]
     for case in results['cases']:
         row = [case['id']]
         for env in results['envs']:
-            row.append(ascii_one(results['tab'][case['id']][env['id']]))
+            res = results['tab'][case['id']][env['id']]
+            if dim is None:
+                dim = res['dimension']
+            else:
+                assert dim == res['dimension']
+            row.append(ascii_one(res))
         tab.append(row)
 
-    return tabulate(tab)
+    return f'All results are in {dim}\n\n' + tabulate(tab)
-- 
2.21.3



  parent reply	other threads:[~2020-10-21 15:09 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-21 14:58 [PATCH v7 00/21] preallocate filter Vladimir Sementsov-Ogievskiy
2020-10-21 14:58 ` [PATCH v7 01/21] block: simplify comment to BDRV_REQ_SERIALISING Vladimir Sementsov-Ogievskiy
2020-10-21 14:58 ` [PATCH v7 02/21] block/io.c: drop assertion on double waiting for request serialisation Vladimir Sementsov-Ogievskiy
2020-10-21 14:58 ` [PATCH v7 03/21] block/io: split out bdrv_find_conflicting_request Vladimir Sementsov-Ogievskiy
2020-10-21 14:58 ` [PATCH v7 04/21] block/io: bdrv_wait_serialising_requests_locked: drop extra bs arg Vladimir Sementsov-Ogievskiy
2020-10-21 14:58 ` [PATCH v7 05/21] block: bdrv_mark_request_serialising: split non-waiting function Vladimir Sementsov-Ogievskiy
2020-10-21 14:58 ` [PATCH v7 06/21] block: introduce BDRV_REQ_NO_WAIT flag Vladimir Sementsov-Ogievskiy
2020-10-21 14:58 ` [PATCH v7 07/21] block: bdrv_check_perm(): process children anyway Vladimir Sementsov-Ogievskiy
2020-11-13 13:59   ` Max Reitz
2020-10-21 14:58 ` [PATCH v7 08/21] block: introduce preallocate filter Vladimir Sementsov-Ogievskiy
2020-11-13 14:32   ` Max Reitz
2020-11-13 15:46     ` Vladimir Sementsov-Ogievskiy
2020-10-21 14:58 ` [PATCH v7 09/21] qemu-io: add preallocate mode parameter for truncate command Vladimir Sementsov-Ogievskiy
2020-10-21 14:58 ` [PATCH v7 10/21] iotests: qemu_io_silent: support --image-opts Vladimir Sementsov-Ogievskiy
2020-10-21 14:58 ` [PATCH v7 11/21] iotests.py: execute_setup_common(): add required_fmts argument Vladimir Sementsov-Ogievskiy
2020-11-13 14:42   ` Max Reitz
2020-10-21 14:58 ` [PATCH v7 12/21] iotests: add 298 to test new preallocate filter driver Vladimir Sementsov-Ogievskiy
2020-11-13 14:43   ` Max Reitz
2020-10-21 14:58 ` [PATCH v7 13/21] scripts/simplebench: fix grammar: s/successed/succeeded/ Vladimir Sementsov-Ogievskiy
2020-11-13 14:44   ` Max Reitz
2020-10-21 14:58 ` Vladimir Sementsov-Ogievskiy [this message]
2020-11-13 15:02   ` [PATCH v7 14/21] scripts/simplebench: support iops Max Reitz
2020-10-21 14:58 ` [PATCH v7 15/21] scripts/simplebench: use standard deviation for +- error Vladimir Sementsov-Ogievskiy
2020-11-13 15:30   ` Max Reitz
2020-10-21 14:58 ` [PATCH v7 16/21] simplebench: rename ascii() to results_to_text() Vladimir Sementsov-Ogievskiy
2020-11-13 15:31   ` Max Reitz
2020-10-21 14:58 ` [PATCH v7 17/21] simplebench: move results_to_text() into separate file Vladimir Sementsov-Ogievskiy
2020-11-13 15:39   ` Max Reitz
2020-10-21 14:58 ` [PATCH v7 18/21] simplebench/results_to_text: improve view of the table Vladimir Sementsov-Ogievskiy
2020-11-13 15:59   ` Max Reitz
2020-11-13 16:24     ` Vladimir Sementsov-Ogievskiy
2020-10-21 14:58 ` [PATCH v7 19/21] simplebench/results_to_text: add difference line to " Vladimir Sementsov-Ogievskiy
2020-10-24 10:24   ` Vladimir Sementsov-Ogievskiy
2020-11-13 16:21   ` Max Reitz
2020-10-21 14:58 ` [PATCH v7 20/21] simplebench/results_to_text: make executable Vladimir Sementsov-Ogievskiy
2020-11-13 16:22   ` Max Reitz
2020-10-21 14:58 ` [PATCH v7 21/21] scripts/simplebench: add bench_prealloc.py Vladimir Sementsov-Ogievskiy
2020-11-13 16:24   ` Max Reitz
2020-11-13 18:00     ` Vladimir Sementsov-Ogievskiy
2020-11-13 19:32       ` Max Reitz
2020-11-13 19:33 ` [PATCH v7 00/21] preallocate filter Max Reitz
2020-11-16  9:32   ` Vladimir Sementsov-Ogievskiy

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20201021145859.11201-15-vsementsov@virtuozzo.com \
    --to=vsementsov@virtuozzo.com \
    --cc=armbru@redhat.com \
    --cc=den@virtuozzo.com \
    --cc=eblake@redhat.com \
    --cc=fam@euphon.net \
    --cc=kwolf@redhat.com \
    --cc=mreitz@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).