|
At 01:06 PM 8/7/00 -0400, Kyle Sparger wrote:
>> How about have n ethernet cards in one machine, could that serve
>> the same purpose?
>
>Still very easily limited, and probably more resource intensive. In most
>PC computers, you'll start having problems once you hit 4 or more ethernet
>cards.
>
>Sending over one ethernet card would be very simple, too. Just dump the
>information to a broad/multi-cast address, and you only have to send one
>packet, rather than 1 packet per destination.
>
>That's not great for security, I know, but it would reduce the time the
>primary LVS server spends on transmitting updates to 1/n of the amount of
>time that transmitting it to n servers would take. Considering that this
>is supposed to be a _HIGH_ volume, very scalable project, I think that the
>conclusion is simple to come to:
>
>On one hand, you have no growth in transmission time whatsoever, on the
>other, you have linear growth. If you're getting 10,000 connections a
>second, with say, 4 potential directors, that's 10,000 updates versus
>40,000. What if you want each server in the cluster to be a potential
>director, and you have 50 of them? Much pain, if you're not
>broad/multi-casting it.
10,000 connection/sec is a huge number. There are not many sites on
the world has that many new connections per second. I was managing
a site had over one million visitors a day in the past, that only got
about 50 new connections per second peak. To get 10,000 new
connection per second, that will take a lot marketing money :) Our tests
showed that with a 733MHz PIII CPU, it is possible to handle that many
new connection/sec on a single LVS box. Alteon and Fundry
are always claiming the number of connection per second they can
handle, but in reality, there is not much need for that. If LVS wants
to handle that much new connection per second, I would think we may
need to take a different approach by modifying the ethernet drivers
to do layer 2 switching than adding more box together.
Wayne
>*shrug*
>
>Thanks,
>
>Kyle Sparger - Senior System Administrator
>Dialtone Internet - Extremely Fast Web Systems
>(954) 581-0097 - Voice (954) 581-7629 - Fax
>ksparger@xxxxxxxxxxxxxxxxxxxx
>http://www.dialtoneinternet.net
>
>On Mon, 7 Aug 2000, Wayne wrote:
>
>> At 12:38 PM 8/7/00 -0400, Joseph Mack wrote:
>> >On Mon, 7 Aug 2000, Wayne wrote:
>> >
>> >> >Parallel port only works for 2 machines. I know you don't want to rely on
>> >> >IP when IP could be hosed, but Ted elects a director from a set of peers.
>> >> >
>> >> >It would be nice to do this on a medium like (ethernet) which is
>> >> >many-to-many.
>> >>
>> >> Sorry for throwing my 2 cents out, but I think LVS box could never
>> >> be the bottle neck for 99.999% situations. If we are thinking balancing
>> >> on the wide area networks, that would be totally different.
>> >
>> >I'm looking for a solution where any of "n" machines can become the
>> >director, where n>2.
>>
>> Joe,
>>
>> How about have n ethernet cards in one machine, could that serve
>> the same purpose?
>>
>> Wayne
>>
>>
>> >Joe
>> >--
>> >Joseph Mack mack@xxxxxxxxxxx
>> >
>> >
>>
>>
>>
>>
|
