Archive for the ‘Load Balancing’ Category

Using ApacheBench to benchmark SSL performance on the Enterprise R16

Monday, September 15th, 2008
Loadbalancer.org are one of a number of vendors that pride themselves on offering affordable load balancing appliances that work. It is the likes of such companies that have collectively driven down the price of these solutions, making load balancing appliances available to companies who previously would not have been in a position to consider such investments.Kemp Technologies are a similar company who’s primary marketing drive centre’s around ‘value for money’. It is because of this glaring similarity that I decided to compare SSL performance capabilities, focusing on the entry-level appliance on offer from each vendor. Specification comparisons were taken, and subsequent performance tests examined whether performance levels met that of the stated specification. Results proved extremely interesting!

(more…)

LVS Local node patch for Linux 2.6.25, Centos 5 kernel build how-to

Monday, July 28th, 2008

Standard Kernel builds of LVS (Linux Vitual Server) don’t have the ability to load balance traffic that is from the local node.
For example if you terminated some SSL traffic using stunnel or pound on the load balancer you then wouldn’t be able to forward that traffic to a backend real server through LVS.

First many thanks to Siim Põder for helping to port Carlos Lozano’s patch from 2.4 -> 2.6
In order to run an SSL reverse proxy on the same node that is running LVS
i.e.

External client —> pound:443 –> Local:443 —> IPVS:80 —> RealServer

The patch for Linux Kernel 2.6.25 is here:
http://www.loadbalancer.org/download/patches/ip_vs_locallvs.patch

The following is a guide how to install on Centos 5.1:
(more…)

Load Balancer Comparison - A Refreshingly Simple Comparison of load balancing hardware specifications.

Monday, May 12th, 2008

“Where can I find good quality Load Balancer information?”

Well that depends if you want it to be biased or not. A lot of our competitors have started setting up dodgy comparison sites highlighting their best features and ignoring the draw backs.

I’m not saying that I’m un-biased when it comes to load balancers I have a pretty strong opinion. But, as far as I’m aware the chart bellow is an accurate comparison of load balancing hardware, price, performance and capability:

(more…)

Why do SSL certificates cost so much?

Friday, May 9th, 2008

I was reading a post by Tony Bourkelicense to SSL“about the licencing restrictions of Verisign et al. when it comes to web sites running on clusters.
He noted a common mis-conception that if you host the SSL cert on the load balancer then you negate the need to pay for one licence per server..
WRONG… you still need to pay for each server in the cluster… wow and I thought it was bad enough to get charged for physicaly copying the cert…

(more…)

Direct Routing aka. Direct Server Return on Windows 2008 using loopback adpter

Wednesday, May 7th, 2008

Direct Routing aka. Direct Server Return (DSR) is a great load balancing method, the idea being that incoming traffic comes into the Virtual IP (VIP) on the load balancer.
Then all the load balancer does is change the destination MAC address of the packet (to one of the destination real servers in the pool) and flips it back to the switch which duefully delivers the packets to the selected real server.

The packet will say “Hello are you the VIP?”

Then the real server will say, “Get lost no I’m not!”.

(more…)

Why Layer 7 load balancing sucks…

Thursday, August 2nd, 2007

Not Sure if if should put a caveat at the top but:  Willy Tarreau has written an excellent argument on behalf of Layer 7 Load Balancers :

Layer 7 is part of the OSI model called the application layer. A typical example would be a web server or database server.

While load balancing hardware marketing execs get very excited about the fact that their product can magically scale your application by using amazing Layer 7 technology in the load balancer such as cookie inserts and tracking/re-writing. What they fail to mention is that any application that requires the load balancer to keep track of session related information within the communications stream can never ever be scalable or reliable.

But lets step back a minute and think about what we are trying to achieve with our load balancing solution.

Are we just looking for increasing the load or performance of our application by adding more application servers?

Or are we trying to achieve true scalability and true horizontal scaling to our application?

(more…)