====================================================================================================
SYSBENCH BENCHMARK REPORT  compared to staging historical data
====================================================================================================
MACHINE:  drizzle-build-n01.wc1.dfw1.stabletransit.com
RUN ID:   735
RUN DATE: 2010-12-15T14:28:16.549501
WORKLOAD: innodb_1000K_readonly
SERVER:   drizzled
VERSION:  lp:~brianaker/drizzle/check-identifiers
REVISION: 1995
COMMENT:  1995: Brian Aker 2010-12-15 [merge] Merge in trunks.
====================================================================================================

TRENDING OVER LAST 5 runs  compared to staging historical data
Conc   TPS     % Diff from Avg   Diff       Min        Max        Avg        
STD       
====================================================================================================
16     1780.36       -2.50%     -45.58    1725.40    1905.63    1825.94      
47.14
32     2307.93       -4.92%    -119.32    2333.37    2481.86    2427.25      
38.80
64     2514.65       +1.70%      42.11    2380.18    2560.61    2472.54      
56.20
128    2471.64       -0.05%      -1.23    2424.10    2545.77    2472.87      
37.22
256    2449.46       +0.21%       5.03    2394.48    2474.97    2444.43      
24.46
512    2222.69       +0.14%       3.11    2189.69    2240.14    2219.58      
15.57
1024   1364.68       -1.02%     -14.08    1332.19    1456.32    1378.76      
30.93
====================================================================================================

TRENDING OVER Last 20 runs  compared to staging historical data

Conc   TPS     % Diff from Avg   Diff       Min        Max        Avg        
STD       
====================================================================================================
16     1780.36       -3.18%     -58.50    1725.40    1906.79    1838.86      
41.02
32     2307.93       -4.03%     -96.82    2230.48    2517.56    2404.75      
67.90
64     2514.65       +3.20%      77.99    2325.45    2560.61    2436.66      
62.98
128    2471.64       +0.05%       1.17    2380.34    2545.77    2470.47      
35.67
256    2449.46       +0.35%       8.45    2394.48    2505.88    2441.01      
26.33
512    2222.69       -0.20%      -4.37    2189.69    2251.93    2227.06      
13.34
1024   1364.68       -0.30%      -4.11    1298.46    1468.18    1368.79      
40.63
====================================================================================================

TRENDING OVER ALL runs  compared to staging historical data

Conc   TPS     % Diff from Avg   Diff       Min        Max        Avg        
STD       
====================================================================================================
16     1780.36      -11.97%    -242.15    1724.98    2233.42    2022.52     
107.71
32     2307.93      -12.97%    -343.84    2010.47    2909.92    2651.77     
142.27
64     2514.65       -6.76%    -182.30    2191.91    2998.80    2696.95     
162.90
128    2471.64       -7.37%    -196.65    2107.33    2911.98    2668.29     
118.92
256    2449.46       -3.21%     -81.32    2174.91    2691.02    2530.78      
86.14
512    2222.69      +17.18%     325.88     968.50    2470.12    1896.81     
499.36
1024   1364.68      +18.09%     209.01     485.25    1866.15    1155.67     
484.99
====================================================================================================
FULL REVISION COMMENTARY:

  1992.5.3: Lee Bieber 2010-12-14 [merge] Merge Stewart - fix bug 655805 - 
HailDB Cursor not resetting this->cursor to clustered index after secondary 
index operations
    1992.10.2: Lee Bieber 2010-12-14 [merge] Merge Stewart - fix bug 655805 - 
HailDB Cursor not resetting this->cursor to clustered index after secondary 
index operations
      1977.3.15: Stewart Smith 2010-12-14 read set IS A LIE. It does *NOT* give 
you what you need. single threaded sysbench now works with HailDB if we ignore 
doing the right thing and just violate the API. arse.
      1977.3.14: Stewart Smith 2010-12-14 fix comment
      1977.3.13: Stewart Smith 2010-12-14 HailDB doDeleteRecord do error 
conversion properly
      1977.3.12: Stewart Smith 2010-12-13 HailDB cursor not reset to clustered 
index.
      1977.3.11: Stewart Smith 2010-12-12 some more correctino of error 
conversion in HailDB
      1977.3.10: Stewart Smith 2010-12-12 improve errors being returned from 
read_row_from_haildb
      1977.3.9: Stewart Smith 2010-12-12 properly translate haildb error code 
in delete_all_rows
      1977.3.8: Stewart Smith 2010-12-12 properly return errors from HailDB 
doUpdateRecord and doDeleteRecord. we weren't using the proper translation 
function before
      1977.3.7: Stewart Smith 2010-12-12 return success/failure from HailDB 
doStartTransaction
    1992.10.1: Lee Bieber 2010-12-14 [merge] Merge Andrew - fix bug 690083: 
Wrong results with regular expressions
      1992.9.1: Andrew Hutchings 2010-12-14 Fix REGEXP which was using the 
match against value for the regex as well
  1992.5.2: Lee Bieber 2010-12-13 [merge] Merge Andrew - fix bug 686773: 
TableFunction::Generator::push() needs assert
    1992.7.3: Lee Bieber 2010-12-13 [merge] Merge Andrew - fix bug 686773: 
TableFunction::Generator::push() needs assert
      1992.8.2: Andrew Hutchings 2010-12-12 More test case fixes
      1992.8.1: Andrew Hutchings 2010-12-12 Enforce 
TableFunction::Generator::push() should only be used on a nullable field
    1992.7.2: Lee Bieber 2010-12-13 [merge] Merge Stewart - ON DUPLICATE KEY 
UPDATE support for HailDB
      1977.3.6: Stewart Smith 2010-12-12 add basic ON DUPLICATE KEY UPDATE test 
for HailDB
      1977.3.5: Stewart Smith 2010-12-10 [merge] merge trunk
      1977.3.4: Stewart Smith 2010-12-10 properly support INSERT ON DUPLICATE 
KEY UPDATE for HailDB engine. Also needed to fix a bug where there was a 
missing move_field_offset(-offset) call in read_row_from_haildb. This could 
have caused all sorts of other bugs too
      1977.3.3: Stewart Smith 2010-12-08 HailDB: make write_row_to_haildb_tuple 
take a parameter for the row buffer to read from
      1977.3.2: Stewart Smith 2010-12-08 basic support for returning index name 
for duplicate key error for HailDB: use ib_get_duplicate_key() to find the 
index name that was violated, then search for the errkey id and return to user.
    1992.7.1: Lee Bieber 2010-12-13 [merge] Merge Monty - fix bug 688819: 
drizzle fails to build from source in natty   
      1992.6.1: Monty Taylor 2010-12-12 Use LT version of lib instead in link 
step
  1992.5.1: Lee Bieber 2010-12-10 [merge] Merge Evan - Add some extra details 
to the libdrizzle documentation
    1992.2.3: Lee Bieber 2010-12-10 [merge] Merge Evan - Add some extra details 
to the libdrizzle documentation
      1992.4.1: Evan Jones 2010-12-10 libdrizzle: Add detail to API docs for 
column and row *_read()
    1992.2.2: Lee Bieber 2010-12-10 [merge] Merge Shrews - Fix compiler warning 
about "comparison between signed and unsigned integer exp
      1992.3.1: David Shrewsbury 2010-12-10 Fix compiler warning in 
transaction_reader code regarding type comparison
    1992.2.1: Lee Bieber 2010-12-10 [merge] Merge Monty - We were leaving 
link-warning.h out of the tarball
      1986.1.1: Monty Taylor 2010-12-09 Distribute link-warning.h and 
distribute it with the right copyright header.

_______________________________________________
Mailing list: https://launchpad.net/~drizzle-benchmark
Post to     : [email protected]
Unsubscribe : https://launchpad.net/~drizzle-benchmark
More help   : https://help.launchpad.net/ListHelp
  • [Drizzle-benchmark] S... drizzle-build-n01 . wc1 . dfw1 . stabletransit . com

Reply via email to