Re: [Boost-users] asio: how to handle disk IO, recommendations

Subject: Re: [Boost-users] asio: how to handle disk IO, recommendations
From: Avi Bahra (avibahra_at_[hidden])
Date: 2012-10-23 12:50:45

> Currently I has a single threaded server based on boost ASIO.
> This works correctly most of the time.
> However on systems where i know that the disk's are slow.
> , the server grinds to a halt, when dealing with the blocking
> disk IO.
>
> I could not find any examples or documentation in ASIO, in
> how to get round these issues. It appears that ASIO does
> not provide any support for async' disk IO.
>
> So I assume this is handled by use of threads ?
> Are there any examples of using ASIO with threaded disk IO,
> since this must be very common use case ?
>
> Is the strand functionality appropriate for handing disk IO,
> in an async' manner.
>
> Any recommendations/advice on this issues would be
> greatly appreciated.
>
> Best regards,
> Ta,
>     Avi
> _______________________________________________
> Boost-users mailing list
> Boost-users_at_[hidden]
> http://lists.boost.org/mailman/listinfo.cgi/boost-users
>

>Hi Avi,
>
>If your server performs short read/writ frequently and data to be transfer
>to network depend on these operations, then multi-threading does not help a
>lot here. I would suggest you to recheck your I/O strategies and make sure
>they are optimized. e.g. buffering, cache, .. etc. If they are perfect but
>performance still bother you, then consider to use memory mapped file or
>libaio.
>
>Best,
>Acer.

In my  case the ASIO server needs to only write to a local disk.(no reads)
One write is relatively short, i.e to a log file, to record all requests,
the other is to write periodically a in memory node tree.
This can vary between 1-100MB.
My thinking was that if I used a thread reserved for writing the
node tree, it will not block the server from handling requests,
 on slow file systems.

I will look into memory mapped files and libaio.

Ta, Avi