|
bobby.moore@xxxxxxxxxxxxx wrote:
>
> My LVS's had a ONE-TIME cost of around $50.00, and access to a broom closet
> full of discarded p133's and p200's. For that $50 I also got 'heartbeat',
> mon, ipchains, iptraf, power tools, a programmable unix environment, a
> plethora of monitoring and benchmarking tools, enough routers, firewalls,
> and virtual servers to potentially put throughout our entire enterprise,
> and TONS of free help and support, not to mention my company's appreciation
> to solutions to some problems that have been dogging us for quite some
> time. I haven't had a throughput problem yet. I haven't stress-tested it
> yet. But my plan is to simply add another LVS and use DNS to round-robin
> them - something we're doing today with another product. To me, the
> potential of this is enormous. I'm starting to see the huge benefit of open
> source. This is not a marketing pitch. Marketing is not my business. There
> is a beauty to LVS, and as a programmer I'd rather have this type product
> to run on this type of system to do what I want with, and produce products
> that will work for me, and to give my 'clients' what they want, rather then
> have them play by my rules. My bent, of course, is that of a programmer and
> not a network guy.
>
> Bobby Moore Worldspan
> Phone: 770.563.7362 Fax: 770.563.6406
> bobby.moore@xxxxxxxxxxxxx
Hi,
I'm evaluating and testing this code since one year now (like the others
during my free time.). I also had to test several commercial products,
like
Localdirector (Cisco), ACEdirector2 (Alteon), ServerIron (Foundry) and
BigIP (F5). What was missing in my little evaluation I did was always
the
flexibility: I couldn't do advanced application level healtchecking. I
could
however do simple http/s-get healthchecks. But with the lvs
implementation
I was able to program whatever the customer wanted me to check in every
language supported under linux. This was by far more important for a
customer
than having a backplane of 9GBit/s. Moreover, when I consider most of
the
sites normally don't provide more then 533Mbit/s through the router. And
then, you are probably running into problems with the 100Mbit ethernet.
We
currently run three HA loadbalancers behind a 155Mbit/s router net with
customers having up to 16 webserver. Why should I put an ACEdirector
switch
being able to transfer in Gbit-range when my network doesn't support it
(which
I suppose could be most of the readers networks too). Second, lvs can be
made HA without major problems, something what sometimes is not an easy
task
for commercial products. Third, they would cost an incredible amount of
money -
ACEdirector would go up to 35000$ for a HA 8 port!! system - So, what if
a
customer wants to add some more servers? You are out of ports, you have
to
cascade with another 8000$. Next, it absolutely no problem to
loadbalance
different customers in different zones with one single loadbalancer
having
multiple ethernet cards or even quadboards and some filter rules. I
actually
programmed such a setup tool in bash with session limitation for VIP and
support for spare server in case every productional server is down. Try
this
with a commercial solution. Moreover, most of the time networking is
poorly
implemented, allowing lots of DoS-attacks. The filter rules are not so
extensible as for example ipchains. I just want to say: I'm extremely
happy
with the lvs because I can extend it as much as I want to.
One more happy user,
Roberto Nibali, ratz
|
