You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@spark.apache.org by Mich Talebzadeh <mi...@gmail.com> on 2018/04/14 19:17:41 UTC
Performance of Spark when the compute and storage are separated
Hi,
This is a sort of your mileage varies type question.
In a classic Hadoop cluster, one has data locality when each node includes
the Spark libraries and HDFS data. this helps certain queries like
interactive BI.
However running Spark over remote storage say Isilon scaled out NAS instead
of LOCAL HDFS becomes problematic. The full-scan Spark needs to do will
take much longer when it is done over the network (access the remote Isilon
storage) instead of local I/O request to HDFS.
Has anyone done some comparative studies on this?
Thanks
Dr Mich Talebzadeh
LinkedIn * https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw
<https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw>*
http://talebzadehmich.wordpress.com
*Disclaimer:* Use it at your own risk. Any and all responsibility for any
loss, damage or destruction of data or any other property which may arise
from relying on this email's technical content is explicitly disclaimed.
The author will in no case be liable for any monetary damages arising from
such loss, damage or destruction.
Re: Performance of Spark when the compute and storage are separated
Posted by vincent gromakowski <vi...@gmail.com>.
Barre metal servers with 2 dedicated clusters (spark and Cassandra) versus
1 cluster with colocation. In both case 10 gbps dedicated network.
Le sam. 14 avr. 2018 à 23:17, Mich Talebzadeh <mi...@gmail.com> a
écrit :
> Thanks Vincent. You mean 20 times improvement with data being local as
> opposed to Spark running on compute nodes?
>
> Dr Mich Talebzadeh
>
>
>
> LinkedIn * https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw
> <https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw>*
>
>
>
> http://talebzadehmich.wordpress.com
>
>
> *Disclaimer:* Use it at your own risk. Any and all responsibility for any
> loss, damage or destruction of data or any other property which may arise
> from relying on this email's technical content is explicitly disclaimed.
> The author will in no case be liable for any monetary damages arising from
> such loss, damage or destruction.
>
>
>
> On 14 April 2018 at 21:06, vincent gromakowski <
> vincent.gromakowski@gmail.com> wrote:
>
>> Not with hadoop but with Cassandra, i have seen 20x data locality
>> improvement on partitioned optimized spark jobs
>>
>> Le sam. 14 avr. 2018 à 21:17, Mich Talebzadeh <mi...@gmail.com>
>> a écrit :
>>
>>> Hi,
>>>
>>> This is a sort of your mileage varies type question.
>>>
>>> In a classic Hadoop cluster, one has data locality when each node
>>> includes the Spark libraries and HDFS data. this helps certain queries like
>>> interactive BI.
>>>
>>> However running Spark over remote storage say Isilon scaled out NAS
>>> instead of LOCAL HDFS becomes problematic. The full-scan Spark needs to
>>> do will take much longer when it is done over the network (access the
>>> remote Isilon storage) instead of local I/O request to HDFS.
>>>
>>> Has anyone done some comparative studies on this?
>>>
>>>
>>> Thanks
>>>
>>>
>>> Dr Mich Talebzadeh
>>>
>>>
>>>
>>> LinkedIn * https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw
>>> <https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw>*
>>>
>>>
>>>
>>> http://talebzadehmich.wordpress.com
>>>
>>>
>>> *Disclaimer:* Use it at your own risk. Any and all responsibility for
>>> any loss, damage or destruction of data or any other property which may
>>> arise from relying on this email's technical content is explicitly
>>> disclaimed. The author will in no case be liable for any monetary damages
>>> arising from such loss, damage or destruction.
>>>
>>>
>>>
>>
>
Re: Performance of Spark when the compute and storage are separated
Posted by Mich Talebzadeh <mi...@gmail.com>.
Thanks Vincent. You mean 20 times improvement with data being local as
opposed to Spark running on compute nodes?
Dr Mich Talebzadeh
LinkedIn * https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw
<https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw>*
http://talebzadehmich.wordpress.com
*Disclaimer:* Use it at your own risk. Any and all responsibility for any
loss, damage or destruction of data or any other property which may arise
from relying on this email's technical content is explicitly disclaimed.
The author will in no case be liable for any monetary damages arising from
such loss, damage or destruction.
On 14 April 2018 at 21:06, vincent gromakowski <
vincent.gromakowski@gmail.com> wrote:
> Not with hadoop but with Cassandra, i have seen 20x data locality
> improvement on partitioned optimized spark jobs
>
> Le sam. 14 avr. 2018 à 21:17, Mich Talebzadeh <mi...@gmail.com>
> a écrit :
>
>> Hi,
>>
>> This is a sort of your mileage varies type question.
>>
>> In a classic Hadoop cluster, one has data locality when each node
>> includes the Spark libraries and HDFS data. this helps certain queries like
>> interactive BI.
>>
>> However running Spark over remote storage say Isilon scaled out NAS
>> instead of LOCAL HDFS becomes problematic. The full-scan Spark needs to
>> do will take much longer when it is done over the network (access the
>> remote Isilon storage) instead of local I/O request to HDFS.
>>
>> Has anyone done some comparative studies on this?
>>
>>
>> Thanks
>>
>>
>> Dr Mich Talebzadeh
>>
>>
>>
>> LinkedIn * https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw
>> <https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw>*
>>
>>
>>
>> http://talebzadehmich.wordpress.com
>>
>>
>> *Disclaimer:* Use it at your own risk. Any and all responsibility for
>> any loss, damage or destruction of data or any other property which may
>> arise from relying on this email's technical content is explicitly
>> disclaimed. The author will in no case be liable for any monetary damages
>> arising from such loss, damage or destruction.
>>
>>
>>
>
Re: Performance of Spark when the compute and storage are separated
Posted by vincent gromakowski <vi...@gmail.com>.
Not with hadoop but with Cassandra, i have seen 20x data locality
improvement on partitioned optimized spark jobs
Le sam. 14 avr. 2018 à 21:17, Mich Talebzadeh <mi...@gmail.com> a
écrit :
> Hi,
>
> This is a sort of your mileage varies type question.
>
> In a classic Hadoop cluster, one has data locality when each node includes
> the Spark libraries and HDFS data. this helps certain queries like
> interactive BI.
>
> However running Spark over remote storage say Isilon scaled out NAS
> instead of LOCAL HDFS becomes problematic. The full-scan Spark needs to
> do will take much longer when it is done over the network (access the
> remote Isilon storage) instead of local I/O request to HDFS.
>
> Has anyone done some comparative studies on this?
>
>
> Thanks
>
>
> Dr Mich Talebzadeh
>
>
>
> LinkedIn * https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw
> <https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw>*
>
>
>
> http://talebzadehmich.wordpress.com
>
>
> *Disclaimer:* Use it at your own risk. Any and all responsibility for any
> loss, damage or destruction of data or any other property which may arise
> from relying on this email's technical content is explicitly disclaimed.
> The author will in no case be liable for any monetary damages arising from
> such loss, damage or destruction.
>
>
>
Re: Performance of Spark when the compute and storage are separated
Posted by Mark Hamstra <ma...@clearstorydata.com>.
Keep forgetting to reply to user list...
On Sun, Apr 15, 2018 at 1:58 PM, Mark Hamstra <ma...@clearstorydata.com>
wrote:
> Sure, data locality all the way at the basic storage layer is the easy way
> to avoid paying the costs of remote I/O. My point, though, is that that
> kind of storage locality isn't necessarily the only way to get acceptable
> performance -- it really does depend heavily on your use case and on your
> performance expectations/requirements. In some cases, it can even be
> acceptable to do query federation between data centers, where some of the
> storage is really remote and the costs to access it are quite high; but if
> you're not doing something like trying to bring over all of the remote
> data, and if you are reusing many times the bit of data that you did bring
> in with the very expensive I/O and then cached, overall performance can be
> quite acceptable.
>
> On Sun, Apr 15, 2018 at 1:46 PM, Mich Talebzadeh <
> mich.talebzadeh@gmail.com> wrote:
>
>> Thanks Mark,
>>
>> I guess this may be broadened to the concept of separate compute from
>> storage. Your point on " ... can kind of disappear after the data is
>> first read from the storage layer." reminds of performing Logical IOs as
>> opposed to Physical IOs. But again as you correctly pointed out on the
>> amount of available cache and concurrency that can saturate the hits on the
>> storage. I personally believe that Data locality helps by avoiding these
>> remote IO calls
>>
>> Dr Mich Talebzadeh
>>
>>
>>
>> LinkedIn * https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw
>> <https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw>*
>>
>>
>>
>> http://talebzadehmich.wordpress.com
>>
>>
>> *Disclaimer:* Use it at your own risk. Any and all responsibility for
>> any loss, damage or destruction of data or any other property which may
>> arise from relying on this email's technical content is explicitly
>> disclaimed. The author will in no case be liable for any monetary damages
>> arising from such loss, damage or destruction.
>>
>>
>>
>> On 15 April 2018 at 21:22, Mark Hamstra <ma...@clearstorydata.com> wrote:
>>
>>> This is a sort of your mileage varies type question.
>>>>
>>>
>>> Yes, it really does. Not only does it depend heavily on the
>>> configuration of your compute and storage, but it also depends a lot on any
>>> caching that you are doing between compute and storage and on the nature of
>>> your Spark queries/Jobs. If you are mostly doing cold full scans, then
>>> you're going to see a big performance hit. If you are reusing a lot of
>>> prior or intermediate results, then you are frequently not going all the
>>> way back to a slow storage layer, but rather to a Spark CachedTable, some
>>> other cache, or even the OS buffer cache for shuffle files -- or to local
>>> disk spillage. All of that is typically going to be local to your compute
>>> nodes, so the data locality issue can kind of disappear after the data is
>>> first read from the storage layer.
>>>
>>>
>>> On Sat, Apr 14, 2018 at 12:17 PM, Mich Talebzadeh <
>>> mich.talebzadeh@gmail.com> wrote:
>>>
>>>> Hi,
>>>>
>>>> This is a sort of your mileage varies type question.
>>>>
>>>> In a classic Hadoop cluster, one has data locality when each node
>>>> includes the Spark libraries and HDFS data. this helps certain queries like
>>>> interactive BI.
>>>>
>>>> However running Spark over remote storage say Isilon scaled out NAS
>>>> instead of LOCAL HDFS becomes problematic. The full-scan Spark needs
>>>> to do will take much longer when it is done over the network (access the
>>>> remote Isilon storage) instead of local I/O request to HDFS.
>>>>
>>>> Has anyone done some comparative studies on this?
>>>>
>>>>
>>>> Thanks
>>>>
>>>>
>>>> Dr Mich Talebzadeh
>>>>
>>>>
>>>>
>>>> LinkedIn * https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw
>>>> <https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw>*
>>>>
>>>>
>>>>
>>>> http://talebzadehmich.wordpress.com
>>>>
>>>>
>>>> *Disclaimer:* Use it at your own risk. Any and all responsibility for
>>>> any loss, damage or destruction of data or any other property which may
>>>> arise from relying on this email's technical content is explicitly
>>>> disclaimed. The author will in no case be liable for any monetary damages
>>>> arising from such loss, damage or destruction.
>>>>
>>>>
>>>>
>>>
>>>
>>
>
Re: Performance of Spark when the compute and storage are separated
Posted by Mich Talebzadeh <mi...@gmail.com>.
Thanks Mark,
I guess this may be broadened to the concept of separate compute from
storage. Your point on " ... can kind of disappear after the data is first
read from the storage layer." reminds of performing Logical IOs as opposed
to Physical IOs. But again as you correctly pointed out on the amount of
available cache and concurrency that can saturate the hits on the storage.
I personally believe that Data locality helps by avoiding these remote IO
calls
Dr Mich Talebzadeh
LinkedIn * https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw
<https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw>*
http://talebzadehmich.wordpress.com
*Disclaimer:* Use it at your own risk. Any and all responsibility for any
loss, damage or destruction of data or any other property which may arise
from relying on this email's technical content is explicitly disclaimed.
The author will in no case be liable for any monetary damages arising from
such loss, damage or destruction.
On 15 April 2018 at 21:22, Mark Hamstra <ma...@clearstorydata.com> wrote:
> This is a sort of your mileage varies type question.
>>
>
> Yes, it really does. Not only does it depend heavily on the configuration
> of your compute and storage, but it also depends a lot on any caching that
> you are doing between compute and storage and on the nature of your Spark
> queries/Jobs. If you are mostly doing cold full scans, then you're going to
> see a big performance hit. If you are reusing a lot of prior or
> intermediate results, then you are frequently not going all the way back to
> a slow storage layer, but rather to a Spark CachedTable, some other cache,
> or even the OS buffer cache for shuffle files -- or to local disk spillage.
> All of that is typically going to be local to your compute nodes, so the
> data locality issue can kind of disappear after the data is first read from
> the storage layer.
>
>
> On Sat, Apr 14, 2018 at 12:17 PM, Mich Talebzadeh <
> mich.talebzadeh@gmail.com> wrote:
>
>> Hi,
>>
>> This is a sort of your mileage varies type question.
>>
>> In a classic Hadoop cluster, one has data locality when each node
>> includes the Spark libraries and HDFS data. this helps certain queries like
>> interactive BI.
>>
>> However running Spark over remote storage say Isilon scaled out NAS
>> instead of LOCAL HDFS becomes problematic. The full-scan Spark needs to
>> do will take much longer when it is done over the network (access the
>> remote Isilon storage) instead of local I/O request to HDFS.
>>
>> Has anyone done some comparative studies on this?
>>
>>
>> Thanks
>>
>>
>> Dr Mich Talebzadeh
>>
>>
>>
>> LinkedIn * https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw
>> <https://www.linkedin.com/profile/view?id=AAEAAAAWh2gBxianrbJd6zP6AcPCCdOABUrV8Pw>*
>>
>>
>>
>> http://talebzadehmich.wordpress.com
>>
>>
>> *Disclaimer:* Use it at your own risk. Any and all responsibility for
>> any loss, damage or destruction of data or any other property which may
>> arise from relying on this email's technical content is explicitly
>> disclaimed. The author will in no case be liable for any monetary damages
>> arising from such loss, damage or destruction.
>>
>>
>>
>
>