[PATCH RFC 00/11] makedumpfile: parallel processing

["Zhou, Wenjian/周文剑"]

On 12/04/2015 04:56 PM, Chao Fan wrote:
> Hi Zhou Wenjian and Kumagai,
>
> I have follow Zhou Wenjian's words to do some tests, in the condition of
> "-c", makdumpfile 1.5.9 does perform better than "-l".
>
> I have done more tests in a machine with 128G memory, in the condition
> of "-d 0" and "-d 3", the makedumpfile 1.5.9 performs well. But if with
> "--num-threads 1", it does need more time than without "--num-threads".
>
> Here is my results(makedumpfile -c):
>
> "-d 0" (the size of vmcore is 2.6G):
> --num-threads        time(seconds)
>      0                 556
>      1                1186
>      4                 307
>      8                 186
>     12                 131
>     16                 123
>
>
> "-d 3" (the size of vmcore is 1.3G):
> --num-threads        time(seconds)
>      0                 141
>      1                 262
>      2                 137
>      4                  91
>      8                 121
>     16                 137
>

Hello Chao,

This result also seems not so good.
We had test it, and you can refer to:
http://lists.infradead.org/pipermail/kexec/2015-October/014576.html

Could you collect the information by *perf stat -e page-faults* on both
--num-threads 0 and --num-threads 1 ?
Your result looks like the performance without the patch which dividing
compress2().

-- 
Thanks
Zhou
> So, I think makedumpfile 1.5.9 can save time in the condition of "-c"
> and not "-d 31" and not "--num-threads 1".
>
> ----- Original Message -----
>> From: "Wenjian Zhou/周文剑" <zhouwj-fnst at cn.fujitsu.com>
>> To: "Atsushi Kumagai" <ats-kumagai at wm.jp.nec.com>
>> Cc: kexec at lists.infradead.org
>> Sent: Friday, December 4, 2015 11:33:36 AM
>> Subject: Re: [PATCH RFC 00/11] makedumpfile: parallel processing
>>
>> Hello Kumagai,
>>
>> On 12/04/2015 10:30 AM, Atsushi Kumagai wrote:
>>> Hello, Zhou
>>>
>>>> On 12/02/2015 03:24 PM, Dave Young wrote:
>>>>> Hi,
>>>>>
>>>>> On 12/02/15 at 01:29pm, "Zhou, Wenjian/周文剑" wrote:
>>>>>> I think there is no problem if other test results are as expected.
>>>>>>
>>>>>> --num-threads mainly reduces the time of compressing.
>>>>>> So for lzo, it can't do much help at most of time.
>>>>>
>>>>> Seems the help of --num-threads does not say it exactly:
>>>>>
>>>>>      [--num-threads THREADNUM]:
>>>>>          Using multiple threads to read and compress data of each page in
>>>>>          parallel.
>>>>>          And it will reduces time for saving DUMPFILE.
>>>>>          This feature only supports creating DUMPFILE in kdump-comressed
>>>>>          format from
>>>>>          VMCORE in kdump-compressed format or elf format.
>>>>>
>>>>> Lzo is also a compress method, it should be mentioned that --num-threads
>>>>> only
>>>>> supports zlib compressed vmcore.
>>>>>
>>>>
>>>> Sorry, it seems that something I said is not so clear.
>>>> lzo is also supported. Since lzo compresses data at a high speed, the
>>>> improving of the performance is not so obvious at most of time.
>>>>
>>>>> Also worth to mention about the recommended -d value for this feature.
>>>>>
>>>>
>>>> Yes, I think it's worth. I forgot it.
>>>
>>> I saw your patch, but I think I should confirm what is the problem first.
>>>
>>>> However, when "-d 31" is specified, it will be worse.
>>>> Less than 50 buffers are used to cache the compressed page.
>>>> And even the page has been filtered, it will also take a buffer.
>>>> So if "-d 31" is specified, the filtered page will use a lot
>>>> of buffers. Then the page which needs to be compressed can't
>>>> be compressed parallel.
>>>
>>> Could you explain why compression will not be parallel in more detail ?
>>> Actually the buffers are used also for filtered pages, it sounds
>>> inefficient.
>>> However, I don't understand why it prevents parallel compression.
>>>
>>
>> Think about this, in a huge memory, most of the page will be filtered, and
>> we have 5 buffers.
>>
>> page1       page2      page3     page4     page5      page6       page7 .....
>> [buffer1]   [2]        [3]       [4]       [5]
>> unfiltered  filtered   filtered  filtered  filtered   unfiltered  filtered
>>
>> Since filtered page will take a buffer, when compressing page1,
>> page6 can't be compressed at the same time.
>> That why it will prevent parallel compression.
>>
>>> Further, according to Chao's benchmark, there is a big performance
>>> degradation even if the number of thread is 1. (58s vs 240s)
>>> The current implementation seems to have some problems, we should
>>> solve them.
>>>
>>
>> If "-d 31" is specified, on the one hand we can't save time by compressing
>> parallel, on the other hand we will introduce some extra work by adding
>> "--num-threads". So it is obvious that it will have a performance
>> degradation.
>>
>> I'm not so sure if it is a problem that the performance degradation is so
>> big.
>> But I think if in other cases, it works as expected, this won't be a problem(
>> or a problem needs to be fixed), for the performance degradation existing
>> in theory.
>>
>> Or the current implementation should be replaced by a new arithmetic.
>> For example:
>> We can add an array to record whether the page is filtered or not.
>> And only the unfiltered page will take the buffer.
>>
>> But I'm not sure if it is worth.
>> For "-l -d 31" is fast enough, the new arithmetic also can't do much help.
>>
>> --
>> Thanks
>> Zhou
>>
>>
>>
>> _______________________________________________
>> kexec mailing list
>> kexec at lists.infradead.org
>> http://lists.infradead.org/mailman/listinfo/kexec
>>