Re: Measuring size of Session object

From: Alvin Bruney [MVP] (vapor)
Date: 06/02/04

  • Next message: John Linn: "Re: Measuring size of Session object"
    Date: Wed, 2 Jun 2004 10:17:24 -0500
    
    

    right,
    session though is very lean. infact the bottle-neck stems from the
    serialization/deserialization of object access instead of bloated memory.
    Note that a caching strategy also includes putting user specific data in
    viewstate as opposed to session. this is a common approach for large sites.
    the downside is that the page load will be a bit slower, but this approach
    is very scalable.

    -- 
    Regards,
    Alvin Bruney
    [ASP.NET MVP http://mvp.support.microsoft.com/default.aspx]
    Got tidbits? Get it here... http://tinyurl.com/27***
    "Roger Kjærnsrød" <kjernsro@online.no> wrote in message 
    news:uda5mZGSEHA.3052@TK2MSFTNGP12.phx.gbl...
    > Thanks for your thoughts on this issue Alvin and I agree that a good 
    > caching
    > strategy is the way to go for common application data. But what I was
    > thinking about here is the individual user data that is different for each
    > user, so caching would not help in this case.
    >
    > And I am not talking about large amounts of data for each user. I was
    > thinking about like max 5K of data per user, but if there is some kind of
    > overhead for each session object that causes each object to be e.g. 20K 
    > this
    > would make a big difference in the total amount of memory usage for 60000
    > simultaneous users (1.2 GB instead of 300Mb).
    >
    >
    >
    > Regards,
    >
    > Roger
    >
    > "Alvin Bruney [MVP]" <vapor at steaming post office> wrote in message
    > news:%23jQq15%23REHA.1216@TK2MSFTNGP10.phx.gbl...
    >> not that i know of.
    >>
    >> memory usage depends on a whole bunch of stuff. making use of a good
    > caching
    >> strategy with scalability in mind is a better alternative than measuring
    >> quantities of objects since the size of an object can change throwing off
    >> your calculations.
    >>
    >> My own tests show that a dataset with 100K rows occupies 100meg of 
    >> memory.
    >> That's entirely useless information since it is system specific and
    > entirely
    >> dependent on the dimensions of the dataset. It should only be used as a
    >> extremely rough idea, about a level up from a wild guess in importance.
    >>
    >> -- 
    >> Regards,
    >> Alvin Bruney
    >> [ASP.NET MVP http://mvp.support.microsoft.com/default.aspx]
    >> Got tidbits? Get it here... http://tinyurl.com/27***
    >> "Roger Kjærnsrød" <kjernsro@online.no> wrote in message
    >> news:%234n7$f9REHA.3348@TK2MSFTNGP09.phx.gbl...
    >> > We are building an application that must be able to handle up to 60000
    >> > simultaneous users.
    >> >
    >> > Therefore I would like to know how much memory each of the users 
    >> > session
    >> > object uses.
    >> >
    >> > Is there an easy way to measure the size of each Session object?
    >> >
    >> >
    >> >
    >>
    >>
    >
    > 
    

  • Next message: John Linn: "Re: Measuring size of Session object"