Re: [PATCH] cfq-iosched: reduce write depth only if sync was delayed

[Jeff Moyer]

Corrado Zoccolo <czoccolo@xxxxxxxxx> writes:

> Hi Jeff,
> On Wed, Dec 9, 2009 at 7:09 PM, Jeff Moyer <jmoyer@xxxxxxxxxx> wrote:
>> Corrado Zoccolo <czoccolo@xxxxxxxxx> writes:
>>
>>> The numbers look good. Now, there is no penalty in having low_latency
>>> set for sequential writes, and just a small penalty for random ones.
>>> The fact that random reads are faster with low_latency set is interesting.
>>> Is the test is running with your patched tiobench (so that the number
>>> of random operations is comparable with sequential ones)?
>>
>> No, I forgot all about that. ÂThe number of random operations defaults
>> to 4000, which is pretty low. ÂI'll re-run the tests with a number
>> comparable to the sequential runs. ÂSorry about that.
>>
> N.P.

low_latency=1

Random Reads
                              File  Blk   Num                   Avg      Maximum      Lat%     Lat%    CPU
Identifier                    Size  Size  Thr   Rate  (CPU%)  Latency    Latency      >2s      >10s    Eff
---------------------------- ------ ----- ---  ------ ------ --------- -----------  -------- -------- -----
2.6.32                        8192  65536   8   19.35 26.10%    74.639     3278.74   0.00000  0.00000    74
2.6.32                        8192  65536  16   20.40 53.07%   135.695     3705.55   0.00000  0.00000    38

Random Writes
                              File  Blk   Num                   Avg      Maximum      Lat%     Lat%    CPU
Identifier                    Size  Size  Thr   Rate  (CPU%)  Latency    Latency      >2s      >10s    Eff
---------------------------- ------ ----- ---  ------ ------ --------- -----------  -------- -------- -----
2.6.32                        8192  65536   8   64.64 342.2%    22.708     5960.22   0.00153  0.00000    19
2.6.32                        8192  65536  16   61.19 663.7%    47.648    11294.56   0.16098  0.00000     9


low_latency=0

Random Reads
                              File  Blk   Num                   Avg      Maximum      Lat%     Lat%    CPU
Identifier                    Size  Size  Thr   Rate  (CPU%)  Latency    Latency      >2s      >10s    Eff
---------------------------- ------ ----- ---  ------ ------ --------- -----------  -------- -------- -----
2.6.32                        8192  65536   8   19.71 25.32%    70.891     4340.92   0.00000  0.00000    78
2.6.32                        8192  65536  16   22.67 58.94%   121.669     6180.49   0.00229  0.00000    38

Random Writes
                              File  Blk   Num                   Avg      Maximum      Lat%     Lat%    CPU
Identifier                    Size  Size  Thr   Rate  (CPU%)  Latency    Latency      >2s      >10s    Eff
---------------------------- ------ ----- ---  ------ ------ --------- -----------  -------- -------- -----
2.6.32                        8192  65536   8   64.69 339.5%    22.651     6770.99   0.00152  0.00000    19
2.6.32                        8192  65536  16   61.94 663.0%    46.962    10719.13   0.12741  0.00000     9


> if you have time, can you also re-run the test changing:
> iosched/fifo_expire_async to 8 ?

Sure.

Cheers,
Jeff
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/