efficiency of including everyone in on a single email or sending abunch of individual emails? - TCP-IP

This is a discussion on efficiency of including everyone in on a single email or sending abunch of individual emails? - TCP-IP ; I'm trying to write a script that'll send out an email to all employees once a month. To do this, there are two possible command sequences that occur to me: MAIL FROM: RCPT TO: RCPT TO: RCPT TO: DATA whatever ...

+ Reply to Thread
Results 1 to 6 of 6

Thread: efficiency of including everyone in on a single email or sending abunch of individual emails?

  1. efficiency of including everyone in on a single email or sending abunch of individual emails?

    I'm trying to write a script that'll send out an email to all
    employees once a month. To do this, there are two possible command
    sequences that occur to me:

    MAIL FROM:
    RCPT TO:
    RCPT TO:
    RCPT TO:
    DATA
    whatever
    QUIT

    ....or...

    MAIL FROM:
    RCPT TO:
    DATA
    whatever
    RSET
    MAIL FROM:
    RCPT TO:
    DATA
    whatever
    RSET
    MAIL FROM:
    RCPT TO:
    DATA
    whatever
    RSET

    It seems to me that of the two, the former would be the most effecient
    - that it'd take the least amount of CPU resources, etc. The only
    problem is... my own testing suggests that that's not the case - that
    it is in fact the latter that's the faster one. My question is...
    why?

    The latter requires you send the DATA multiple times. If you're
    sending an email to 100 people, that's atleast 100x the amount of
    bandwidth that the latter would need that the former wouldn't...

  2. Re: efficiency of including everyone in on a single email or sendinga bunch of individual emails?

    On Jul 28, 11:02*am, yawnmoth wrote:

    > It seems to me that of the two, the former would be the most effecient
    > - that it'd take the least amount of CPU resources, etc. *The only
    > problem is... *my own testing suggests that that's not the case - that
    > it is in fact the latter that's the faster one. *My question is...
    > why?


    I don't see how you could have measured and not know. You have the
    test data, so you should know why it was faster.

    My bet is that the mail server started to tarpit you. That is, after
    each target was accepted, it forced a longer and longer pause before
    it would let you send the next one.

    DS

  3. Re: efficiency of including everyone in on a single email or sending a bunch of individual emails?

    On Mon, 28 Jul 2008 11:02:06 -0700 (PDT), yawnmoth
    wrote:

    >I'm trying to write a script that'll send out an email to all
    >employees once a month. To do this, there are two possible command
    >sequences that occur to me:


    Why aren't you using an Email-Alias on the Email-Server?

  4. Re: efficiency of including everyone in on a single email or sendinga bunch of individual emails?

    On Jul 28, 7:02 pm, yawnmoth wrote:
    > I'm trying to write a script that'll send out an email to all
    > employees once a month. To do this, there are two possible command
    > sequences that occur to me:
    >
    > MAIL FROM:
    > RCPT TO:
    > RCPT TO:
    > RCPT TO:
    > DATA
    > whatever
    > QUIT
    >
    > ...or...
    >
    > MAIL FROM:
    > RCPT TO:
    > DATA
    > whatever
    > RSET
    > MAIL FROM:
    > RCPT TO:
    > DATA
    > whatever
    > RSET
    > MAIL FROM:
    > RCPT TO:
    > DATA
    > whatever
    > RSET
    >
    > It seems to me that of the two, the former would be the most effecient
    > - that it'd take the least amount of CPU resources, etc. The only
    > problem is... my own testing suggests that that's not the case - that
    > it is in fact the latter that's the faster one.

    How did you time it from start of connect to confirmation from server
    that ALL emails were sent out (I hope)?
    > My question is... why?

    Depends on how you connect to the server, what are the server settings
    and many other things...
    >
    > The latter requires you send the DATA multiple times. If you're
    > sending an email to 100 people, that's atleast 100x the amount of
    > bandwidth that the latter would need that the former wouldn't...

    If you answer the question about how you timed it we may be able to
    help...


  5. Re: efficiency of including everyone in on a single email or sending abunch of individual emails?

    In article <79bd6f30-98d3-46c0-83e0-d6857f5903ea@i76g2000hsf.googlegroups.com>,
    yawnmoth wrote:
    >I'm trying to write a script that'll send out an email to all
    >employees once a month. To do this, there are two possible command
    >sequences that occur to me:
    >
    > ... [sequences snipped] ...
    >
    >It seems to me that of the two, the former would be the most effecient
    >- that it'd take the least amount of CPU resources, etc. The only
    >problem is... my own testing suggests that that's not the case - that
    >it is in fact the latter that's the faster one. My question is...
    >why?
    >
    >The latter requires you send the DATA multiple times. If you're
    >sending an email to 100 people, that's atleast 100x the amount of
    >bandwidth that the latter would need that the former wouldn't...


    With minimal details this is mostly a guess but, whats probably
    happening is with each MAIL FROM, RCPT TO, DATA thats accepted and
    queued by your relay it can proceed to send it before your script is
    finished.

    --
    -- Rod --
    rodd(at)polylogics(dot)com

  6. Re: efficiency of including everyone in on a single email or sendinga bunch of individual emails?

    On Jul 28, 7:05*pm, David Schwartz wrote:
    > On Jul 28, 11:02*am, yawnmoth wrote:
    >
    > > It seems to me that of the two, the former would be the most effecient
    > > - that it'd take the least amount of CPU resources, etc. *The only
    > > problem is... *my own testing suggests that that's not the case - that
    > > it is in fact the latter that's the faster one. *My question is...
    > > why?

    >
    > I don't see how you could have measured and not know. You have the
    > test data, so you should know why it was faster.

    The former takes several hours to complete when being sent to 20,000+
    people, whereas the latter takes maybe ten minutes. I don't have
    access to the SMTP server, so I can't perform tests on that - only
    client side tests.

+ Reply to Thread