Re: [tlug] Making better use of SSDs?

[satoshi.nagayasu@example.com]

Hi,

2012/5/28 Raymond Wan <rwan.kyoto@example.com>:
> Hi all,
>
> My employer is thinking of purchasing a server and wants to also buy
> an SSD along with it.  I can only think of two ways of making use of
> an SSD under Linux and I was wondering if anyone has better ideas.
>
> 1)  Well, make it into a drive.  This would only benefit programs that
> have a lot of I/O or programs like Unix's 'sort' which can allow you
> to specify which temporary directory to use.  I can just specify the
> SSD drive.
>
> 2)  Make it into a swap partition.  This would help if the amount of
> memory used by a program exceeds main memory and it starts paging to
> disk.  Paging to an SSD would surely be faster.  (Assuming it pages
> often...)
>
> The problem with #2 is that you don't see the benefit of the SSD until
> your program thrashes.  And, of course, the SSD still isn't main
> memory.  What's on the SSD will be swapped out static pages.
>
> As an example, suppose the server has at most 32 GB and a 32 GB SSD is
> purchased.  Option #2 is better than not having the SSD at all (of
> course!) but also not the same as having a machine with 64 GB of main
> memory.  :-)  Option #1 relies on running programs that have either a
> lot of I/O or have a way of specifying alternate temporary
> directories.
>
> Can anyone think of a third option?

I recommend #1 option, and have your own experiments to examine
whether SSD benefits your system performance.

A few days ago, I just published a technical report that examine
SSD performance compared with HDD to execute DWH (or analytics)
workloads on PostgreSQL. SSD was used for its data directory.

According to the experiments, it's been determined that SSD is able to
improve the query performance extremely if it has massive random
acccess I/O operations. As a results, more than 20 times faster
performance has been observed with some workloads.

On the other hand, by replacing HDD with SSD, sequential access (read)
was not improved as much as I expected (almost x1.3 faster than HDD),
because sequential access on HDD is enough fast.

Generally speaking, it's safe to say SSD would increase the performance.
But practically, you need to determine a type of workload (or a bottleneck
of workload) that you're running. And it's really fun! :)

Regards,

>
> Thank you!
>
> Ray
>
> --
> To unsubscribe from this mailing list,
> please see the instructions at http://lists.tlug.jp/list.html
>
> The TLUG mailing list is hosted by ASAHI Net, provider of mobile and
> fixed broadband Internet services to individuals and corporations.
> Visit ASAHI Net's English-language Web page: http://asahi-net.jp/en/



-- 
NAGAYASU Satoshi <satoshi.nagayasu@example.com>