Site Sponsors:
Server Setup on CentOS Minimalist 

Back ... to the Future?

Those who have been using Linux & other POSIX systems for a few decades know that there was a time when there was no GUI. While technologies like XWindows are indeed welcome, some times we want to to make every byte & cycle count. -Why waste the floating-point time and space when graphical users are not part of our port-use paradigm?

Serving-up maximum power is surely the case whenever we set up a Server. Greedly managing every performance aspect is also the reason why many love the minimalist installation option available with CentOS.

Get Networking on Minimalist

Of course, the 'minimalist' definition means slightly different things to slightly different server-roles. --Yet since the ability to pick-and-choose is what we are after, here are a few general notes on how to get things started:

ifconfig eth0 netmask up
ip addr add dev eth0
ip route add default via
The example above is where your minimalist VM / server is a Class C (24 bit) net-mask with an IP address of *.231, and *.1 is your Gateway.

Getting Downloads on Minimalist

After you can do things like ping the outside world, the next thing to do is to get wget:

yum install wget

Downloading Apache Derby on Minimalist

Thereafter, my mission was to set up a dedicated Derby server for a custom Advanced Java Developer, Hibernate, Google Earth & "Big Data" Gig. Since license support prohibits a clear wget example of installing JEE / Oracle Java, lets begin by downloading something from the Apache Commons:
(other mirrors here)

Archive Support on Minimalist (Derby Example)

Next, we will all surely want to unzip archives, so:
gunzip *.gz

Such will therefore allow us to obviously:
tar -xvf *.tar
cd db*

so we can get ready for the next steps of installing Java, as well as the Derby SQL Server proper.

Adding Java

When it comes time to install java, we must resist the temptation to use yum. Instead, we should probably simply cross-load our favourite Java installation archive to a handy personal web server, then use wget to entice it down from there. (Surely a tad more involved, new folks should nevertheless neuron-in on the idea that things like FTP are obviously also an option for cross-loading things like "agree first" archives.)

File Wiring

Using our .bashrc is the best way to test things:
export JAVA_HOME=/myinstall/jre
export DERBY_HOME=/myinstall/derby
export CLASSPATH=$DERBY_HOME/lib/derbytools.jar:$DERBY_HOME/lib/derbynet.jar
export PATH="$JAVA_HOME/bin:$DERBY_HOME/bin:$PATH"
Ultimately however, servers need to start other things whenever they start up. Therefore, after updating the default path to where things are for the entire server (vi /etc/profile), be sure to update your list of inittab applications (e.g. startNetworkServer) as well.
startNetworkServer -h
You might even prefer to use the chkconfig utility.


Lest we forget, be sure to allows other computers to connect to your new server by updating your security & port requirements in /etc/sysconfig/iptables, too!

Here is a decent iptables R&D packet filtering stance for an up-and-down VM - such as what one might like to train on:
iptables -F
iptables -X
iptables -t nat -F
iptables -t nat -X
iptables -t mangle -F
iptables -t mangle -X
iptables -P INPUT ACCEPT
Once defined, add the rules to the /etc/sysconfig/iptables file by typing:
/sbin/service iptables save



p.s: If you would like to see a better color choice (i.e. so students can keep track of which VM they are using) then something like:

setterm -background white -foreground blue -store

-always helps! (you might even want to use things like
-blink on
to brighten things up a bit...)

[ view entry ] ( 1836 views )   |  permalink  |  related link
Linkedin: How can my code be faster than STL algorithms?  
Linkedin asks: "Something is wrong and I don't know what. How can my code be faster than STL algorithms?"

My Response:

"Many often find that custom code creates faster solutions. In general, generic solutions produce generic results?

Here your test cases work well for what you have envisioned. Yet if you want to see a REAL performance boost, then pre-allocate your <T> memory for the anticipated usage, only to grow by similarly-sized blocks; Base allocations upon a rounded-up, architecture specific (64 bit?) sub-allocated power of 2. Thereafter, if the preferred usage is required to be full CRUD, consider using a linked list and a pack() scenario for sub reallocation and no one will be able to touch your speed.

For even faster results, from there we might use inline assembly language so as to base those blocks upon Intel Page Selectors... Why? Because calling 'new' all the time is always a performance bottleneck. Using Intel page selectors directly for memory sub-allocation will be allot f-a-s-t-e-r. --But is it worth the effort?

Get the idea? Generic solutions can almost always be improved upon by by custom means / coding to support a particular problem domain. Many times – when screaming performance is required - it can even be better to simply junk generics all together... It all depends upon what one is trying to accomplish?

[ view entry ] ( 2322 views )   |  permalink  |  related link
Linkedin: Working Less for Better Results? 
My reply to a post on linkedin:

"Surely - as noted - a good part of strategic thinking often involves simply considering what others are doing / have done. Yet rather than working less, many often find that 'never leaving work' is part and parcel of what most employers are looking for in a professional. (i.e. If we need a doctor, do we not want one that keeps up with his or her trade in their spare-time?)

Likewise, the willingness to understand the patterns and practices from elsewhere often requires a gift of insight that only repeated exposure / constant work can provide. -Such is why folks need Consultants. Surely many have also discovered that looking for bigger-picture lessons often requires a great deal more - rather than less - work.

Yet when we truly enjoy what we are doing, the extra efforts often seems like no-work at-all?

So I do not believe that one should ever look at strategic assessments as a 9-5 activity. If one is genuinely a professional, then strategic thinking takes place far MORE - rather than less - frequently?

Yet perhaps the best way to be ever thinking about one's profession is to simply be sure that one has chosen a profession that one enjoys. ...

"The biggest mistake most people make is trying to earn a living doing something that they do not enjoy." (sic) ---Forbes

[ view entry ] ( 1215 views )   |  permalink  |  related link

<<First <Back | 41 | 42 | 43 | 44 | 45 | 46 | 47 | 48 | 49 | 50 | Next> Last>>