You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@cassandra.apache.org by Gabriel Menegatti <ga...@s1mbi0se.com.br> on 2015/02/01 21:32:08 UTC
Any problem mounting a keyspace directory in ram memory?
Hi guys,
Please, does anyone here already mounted a specific keyspace directory to
ram memory using tmpfs?
Do you see any problem doing so, except by the fact that the data can be
lost?
Thanks in advance.
Regards,
Gabriel.
Re: Any problem mounting a keyspace directory in ram memory?
Posted by Hannu Kröger <hk...@gmail.com>.
At least I cannot think of any reason why it wouldn't work. As you said, you might lose the data but if you can live with that then why not.
Hannu
> On 02.02.2015, at 14:21 , Gabriel Menegatti <ga...@s1mbi0se.com.br> wrote:
>
> Hi Colin,
>
> Yes, we don't want to use the C* in-memory, we just want to mount the keyspace data directory to RAM instead of leaving it on the spinning disks.
>
> My question is more related to the technical side of mounting the keyspace data folder to the ram memory than checking if Cassandra has some in-memory feature.
>
> My intention is to understand if mounting a keyspace data directory to RAM could cause any technical problems... On our point of it shouldn't, as we are just "moving" the directory to be stored on RAM instead of the spinning disks.
>
> Thanks so much.
>
>
> Em 02/02/2015, às 05:15, Colin <colin@clark.ws <ma...@clark.ws>> escreveu:
>
>> Until the in-memory option stores data off heap, I would strongly recommend staying away from this option. This was a marketing driven hack in my opinion.
>>
>> --
>> Colin Clark
>> +1 612 859 6129
>> Skype colin.p.clark
>>
>> On Feb 2, 2015, at 5:31 AM, Jan <cnet62@yahoo.com <ma...@yahoo.com>> wrote:
>>
>>> HI Gabriel;
>>>
>>> I don't think Apache Cassandra supports in-memory keyspaces.
>>> However Datastax Enterprise does support it.
>>>
>>> Quoting from Datastax:
>>> DataStax Enterprise includes the in-memory option for storing data to and accessing data from memory exclusively. No disk I/O occurs. Consider using the in-memory option for storing a modest amount of data, mostly composed of overwrites, such as an application for mirroring stock exchange data. Only the prices fluctuate greatly while the keys for the data remain relatively constant. Generally, the table you design for use in-memory should have the following characteristics:
>>> Store a small amount of data
>>> Experience a workload that is mostly overwrites
>>> Be heavily trafficked
>>> Using the in-memory option | DataStax Enterprise 4.0 Documentation <http://www.datastax.com/documentation/datastax_enterprise/4.0/datastax_enterprise/inMemory.html>
>>>
>>>
>>>
>>>
>>>
>>>
>>> Using the in-memory option | DataStax Enterprise 4.0 Documentation
>>> <http://www.datastax.com/documentation/datastax_enterprise/4.0/datastax_enterprise/inMemory.html>Using the in-memory option
>>> View on www.datastax.com <http://www.datastax.com/documentation/datastax_enterprise/4.0/datastax_enterprise/inMemory.html>
>>> Preview by Yahoo
>>>
>>>
>>>
>>> hope this helps
>>> Jan
>>>
>>> C* Architect
>>>
>>>
>>>
>>> On Sunday, February 1, 2015 1:32 PM, Gabriel Menegatti <gabriel@s1mbi0se.com.br <ma...@s1mbi0se.com.br>> wrote:
>>>
>>>
>>> Hi guys,
>>>
>>> Please, does anyone here already mounted a specific keyspace directory to ram memory using tmpfs?
>>>
>>> Do you see any problem doing so, except by the fact that the data can be lost?
>>>
>>> Thanks in advance.
>>>
>>> Regards,
>>> Gabriel.
>>>
>>>
Re: Any problem mounting a keyspace directory in ram memory?
Posted by Gabriel Menegatti <ga...@s1mbi0se.com.br>.
Hi Colin,
Yes, we don't want to use the C* in-memory, we just want to mount the keyspace data directory to RAM instead of leaving it on the spinning disks.
My question is more related to the technical side of mounting the keyspace data folder to the ram memory than checking if Cassandra has some in-memory feature.
My intention is to understand if mounting a keyspace data directory to RAM could cause any technical problems... On our point of it shouldn't, as we are just "moving" the directory to be stored on RAM instead of the spinning disks.
Thanks so much.
> Em 02/02/2015, às 05:15, Colin <co...@clark.ws> escreveu:
>
> Until the in-memory option stores data off heap, I would strongly recommend staying away from this option. This was a marketing driven hack in my opinion.
>
> --
> Colin Clark
> +1 612 859 6129
> Skype colin.p.clark
>
>> On Feb 2, 2015, at 5:31 AM, Jan <cn...@yahoo.com> wrote:
>>
>> HI Gabriel;
>>
>> I don't think Apache Cassandra supports in-memory keyspaces.
>> However Datastax Enterprise does support it.
>>
>> Quoting from Datastax:
>> DataStax Enterprise includes the in-memory option for storing data to and accessing data from memory exclusively. No disk I/O occurs. Consider using the in-memory option for storing a modest amount of data, mostly composed of overwrites, such as an application for mirroring stock exchange data. Only the prices fluctuate greatly while the keys for the data remain relatively constant. Generally, the table you design for use in-memory should have the following characteristics:
>> Store a small amount of data
>> Experience a workload that is mostly overwrites
>> Be heavily trafficked
>> Using the in-memory option | DataStax Enterprise 4.0 Documentation
>>
>>
>>
>>
>>
>>
>> Using the in-memory option | DataStax Enterprise 4.0 Documentation
>> Using the in-memory option
>> View on www.datastax.com
>> Preview by Yahoo
>>
>>
>>
>> hope this helps
>> Jan
>>
>> C* Architect
>>
>>
>>
>> On Sunday, February 1, 2015 1:32 PM, Gabriel Menegatti <ga...@s1mbi0se.com.br> wrote:
>>
>>
>> Hi guys,
>>
>> Please, does anyone here already mounted a specific keyspace directory to ram memory using tmpfs?
>>
>> Do you see any problem doing so, except by the fact that the data can be lost?
>>
>> Thanks in advance.
>>
>> Regards,
>> Gabriel.
>>
>>
Re: Any problem mounting a keyspace directory in ram memory?
Posted by Colin <co...@clark.ws>.
Until the in-memory option stores data off heap, I would strongly recommend staying away from this option. This was a marketing driven hack in my opinion.
--
Colin Clark
+1 612 859 6129
Skype colin.p.clark
> On Feb 2, 2015, at 5:31 AM, Jan <cn...@yahoo.com> wrote:
>
> HI Gabriel;
>
> I don't think Apache Cassandra supports in-memory keyspaces.
> However Datastax Enterprise does support it.
>
> Quoting from Datastax:
> DataStax Enterprise includes the in-memory option for storing data to and accessing data from memory exclusively. No disk I/O occurs. Consider using the in-memory option for storing a modest amount of data, mostly composed of overwrites, such as an application for mirroring stock exchange data. Only the prices fluctuate greatly while the keys for the data remain relatively constant. Generally, the table you design for use in-memory should have the following characteristics:
> Store a small amount of data
> Experience a workload that is mostly overwrites
> Be heavily trafficked
> Using the in-memory option | DataStax Enterprise 4.0 Documentation
>
>
>
>
>
>
> Using the in-memory option | DataStax Enterprise 4.0 Documentation
> Using the in-memory option
> View on www.datastax.com
> Preview by Yahoo
>
>
>
> hope this helps
> Jan
>
> C* Architect
>
>
>
> On Sunday, February 1, 2015 1:32 PM, Gabriel Menegatti <ga...@s1mbi0se.com.br> wrote:
>
>
> Hi guys,
>
> Please, does anyone here already mounted a specific keyspace directory to ram memory using tmpfs?
>
> Do you see any problem doing so, except by the fact that the data can be lost?
>
> Thanks in advance.
>
> Regards,
> Gabriel.
>
>
Re: Any problem mounting a keyspace directory in ram memory?
Posted by Gabriel Menegatti <ga...@s1mbi0se.com.br>.
Hi Jan,
Thanks for your reply, but C* in-memory just supports 1 GB keyspaces at the moment, what is not enough for us.
My question is more related to the technical side of mounting the keyspace data folder to the ram memory than checking if Cassandra has some in-memory feature.
My intention is to understand if mounting a keyspace data directory to RAM could cause any technical problems... On our point of it shouldn't, as we are just "moving" the directory to be stored on RAM instead of in spinning disks.
Thanks so much.
> Em 02/02/2015, às 02:31, Jan <cn...@yahoo.com> escreveu:
>
> HI Gabriel;
>
> I don't think Apache Cassandra supports in-memory keyspaces.
> However Datastax Enterprise does support it.
>
> Quoting from Datastax:
> DataStax Enterprise includes the in-memory option for storing data to and accessing data from memory exclusively. No disk I/O occurs. Consider using the in-memory option for storing a modest amount of data, mostly composed of overwrites, such as an application for mirroring stock exchange data. Only the prices fluctuate greatly while the keys for the data remain relatively constant. Generally, the table you design for use in-memory should have the following characteristics:
> Store a small amount of data
> Experience a workload that is mostly overwrites
> Be heavily trafficked
> Using the in-memory option | DataStax Enterprise 4.0 Documentation
>
>
>
>
>
>
> Using the in-memory option | DataStax Enterprise 4.0 Documentation
> Using the in-memory option
> View on www.datastax.com
> Preview by Yahoo
>
>
>
> hope this helps
> Jan
>
> C* Architect
>
>
>
> On Sunday, February 1, 2015 1:32 PM, Gabriel Menegatti <ga...@s1mbi0se.com.br> wrote:
>
>
> Hi guys,
>
> Please, does anyone here already mounted a specific keyspace directory to ram memory using tmpfs?
>
> Do you see any problem doing so, except by the fact that the data can be lost?
>
> Thanks in advance.
>
> Regards,
> Gabriel.
>
>
Re: Any problem mounting a keyspace directory in ram memory?
Posted by Jan <cn...@yahoo.com>.
HI Gabriel;
I don't think Apache Cassandra supports in-memory keyspaces. However Datastax Enterprise does support it.
Quoting from Datastax: DataStax Enterprise includes the in-memory option for storing data to and accessing data from memory exclusively. No disk I/O occurs. Consider using the in-memory option for storing a modest amount of data, mostly composed of overwrites, such as an application for mirroring stock exchange data. Only the prices fluctuate greatly while the keys for the data remain relatively constant. Generally, the table you design for use in-memory should have the following characteristics:
- Store a small amount of data
- Experience a workload that is mostly overwrites
- Be heavily trafficked
Using the in-memory option | DataStax Enterprise 4.0 Documentation
| |
| | | | | |
| Using the in-memory option | DataStax Enterprise 4.0 DocumentationUsing the in-memory option |
| |
| View on www.datastax.com | Preview by Yahoo |
| |
| |
hope this helpsJan
C* Architect
On Sunday, February 1, 2015 1:32 PM, Gabriel Menegatti <ga...@s1mbi0se.com.br> wrote:
Hi guys,
Please, does anyone here already mounted a specific keyspace directory to ram memory using tmpfs?
Do you see any problem doing so, except by the fact that the data can be lost?
Thanks in advance.
Regards,Gabriel.