As promised, here are the slides from the Understanding Functional Programming session:

If you have any questions or comments, please send Chris an email; he loves to explore and discuss technology.

What topic(s) would you like to see next time?

“Code Camps have been taking place all over the country. This is a free, one-day event put on by the local Phoenix community to help promote software development in general. There is no right or wrong language, platform, or technology. If a topic relates in any way to the code that causes a machine to produce a desired result, it’s welcome here.”

This year it will be held at the DeVry University Phoenix Campus and is once again being organized by Lorin Thwaits.

Flatterline will be teaching two sessions this year in the 21st Century Software Development track: Developing with CouchDB and Understanding functional programming.

The schedule became available today, so if you haven’t registered do so now and build your schedule for Saturday. It looks like there are a lot of great sessions this year and we can’t wait to learn from all of our colleagues. See you there!

Photo by tonx.

We had a great turnout for the first Phoenix OpenCoffee Club, which took place April 30th at Fair Trade Cafe. About a dozen people, both entrepreneurs and investors, attended and spent several hours getting to know each other over some great coffee.

We heard at least 6 ideas pitched and discussed during the 2 hours we were there. The assembled group wasn’t shy about asking questions and raising concerns of the entrepreneurs pitching their ideas either. I know everyone appreciated hearing the perspectives of entrepreneurs and investors alike. We hope everyone who attended found the meetup engaging and useful. We sure did!

Special thanks to Fair Trade Cafe for allowing us to host our event there. The May 7th Phoenix OpenCoffee Club will be held at Hob Nobs Coffee House. We’ll try to have OpenCoffee Club every Thursday, so it’s not a big deal if there are times you can’t make it. Hope to see you there!

A happy application in the cloud.
Photo by Timothy K. Hamilton.

Flatterline believes heavily in the power of cloud computing. Dynamically allocated hardware on a pay-for-what-you-need basis has tremendous advantages when it comes to helping clients manage and provision their clusters. The main advantage of cloud computing is dynamically growing, or shrinking, hardware as the needs of the application change.

Because of the dynamic nature of cloud computing, we don’t need a guaranteed answer on hardware requirements up front. However, a client may want a ballpark figure in order to set aside the right amount of budget or let investors know the estimated operational cost. You could crunch the numbers yourself, but why would you do that when we’ve already automated the process for you?

Determining optimal cost

We’ve constructed a very basic model for minimizing the cost of Amazon EC2 hardware resources which satisfies a minimum number of EC2 Compute Units and a given amount of RAM per process. The technique uses linear programming and the GNU linear programming kit (GLPK). Note: I’m a math geek that likes linear modeling, so if you’re unfamiliar with either, I’d be happy to chat with you about them over lunch.

Installation

First, install the GLPK. On Ubuntu execute the command

sudo aptitude install glpk

on Mac OS X execute the command

sudo port install glpk

Next, download the following gist as cloud_cost.txt.

Computing the cost

The model requires the specification of two variables: total number of EC2 Compute Units and RAM. Both variables are specified at the bottom with param unitsNeeded and param ramRequiredPerAppInstance respectively. Change these params to reflect your particular situation. Note: A future article will explore capacity planning in more detail.

When you’re ready, execute the solver using the following command:

glpsol --model cloud_cost.txt --output result.txt

Analyzing the results

The program generates the result into a file called result.txt. Assuming 500 EC2 Compute Units with 125MB of RAM per process, the file will look something like the following:

Problem:    cloud_cost
Rows:       7
Columns:    5 (5 integer, 0 binary)
Non-zeros:  15
Status:     INTEGER OPTIMAL
Objective:  cost = 14400 (MINimum)

 1 InstanceQuantity[Small]
                    *              0             0
 2 InstanceQuantity[Large]
                    *              0             0
 3 InstanceQuantity[XLarge]
                    *              0             0
 4 InstanceQuantity[HCPULarge]
                    *              0             0
 5 InstanceQuantity[HCPUXLarge]
                    *             25             0

The objective function was cost, so the optimal arrangement of hardware needed to get that computational power costs $14,400/month. The second column of the hardware arrangement indicates the number of instance needed. In this case we need 25 high-CPU, extra large instances.

Did you find this useful? Let us know in the comments!

At Flatterline, our stated mission is to help startups realize their vision. To that end, we’ve decided to help bring OpenCoffee Club to Phoenix.

What is OpenCoffee Club?

OpenCoffee Club is a weekly gathering of entrepreneurs and investors in an informal setting (a.k.a, a coffee shop). The goal is to bring these entrepreneurs and investors together to talk about projects, collaborate and get to know each other.

It was originally started in London by Saul Klein, a venture partner at Index Ventures. Since then, over 80 OpenCoffee Clubs have sprung up worldwide. You can read more about it on Saul’s blog or at ReadWriteWeb.

Why?

On a recent trip to Boulder, CO I attended an OpenCoffee Club facilitated by Jason Mendelson of the Foundry Group. It was really inspiring to see such a vibrant startup community meeting regularly to discuss ideas, share problems, give advice and, most of all, listen to each other.

We believe that Phoenix has a lot of talented, entrepreneurial people. We want them to have the best chance to grow and succeed and this is one small way we see that happening.

The first meetup will be on Thursday, April 30 from 8am – 10am at Fair Trade Coffee. Please let us know you’re coming.

In what other ways can we help the Phoenix startup community thrive?

If you are a highly mobile laptop user, chances are you work out of a lot of varying public locations such as coffee shops, libraries, and just about anywhere that has public wifi. Are you concerned about the privacy of your data? If you’re like me, and our clients, you’re very concerned. Applications like tcpflow and Wireshark are not only particularly effective at grabbing content from the network, they also happen to be readily available.

Here’s a very simple scheme to leverage the inexpensive power of Amazon’s EC2 to create a disposable, secure proxy.

Getting started

You will need the following to make this recipe work:

• Amazon AWS account
• A Ubuntu-based Amazon AMI with keypair (we are using public ami-7cfd1a15 for this article)
• An EC2 security group allowing a minimum of port 22 for SSH

To start, launch a small instance of your AMI of choice.  Once again, we prefer Ubuntu so most of this article is going to be Ubuntu-centric.  This instance will need to be setup with whatever key pair you plan on using as well as be placed in the security group that allows SSH access.  If you need help with this the Amazon AWS console is particularly useful.

Putting the pieces together

Once the instance is made available ssh to your newly created instance.

ssh -i identity_file -L 3128:localhost:3128 root@public_ec2_domain_name

The noteworthy addition to the previous line is -L 3128:localhost:3128. This addition to the SSH command will open port 3128 locally and forward all traffic to the remote port 3128 across the open SSH connection.

Once the connection is open you will need to install a proxy, we prefer squid.  Squid can be installed through the following command:

aptitude update
aptitude install squid

The last remaining step is to configure your browser of choice to use proxy localhost:3128.

The final result is all local HTTP traffic will be relayed across local port 3128 across the encrypted SSH tunnel to the 10 cents / hour remote server. From there it will go out to the internet at large away from the prying eyes of nefarious coffee shop patrons. Keep the SSH connection open for as long as you need access to the proxy.

Cleaning up after yourself

When your done at the coffee shop feel free to decommission the AMI instance and you’re done. The machine will go away with all records of the proxy’s cache.

If you’re looking for a solution to more than just your HTTP traffic you have options such as OpenVPN. Look for an article from us soon.

Swappiness

The Linux 2.6 kernel has a parameter called vm.swappiness that regulates the kernel’s likelihood to swap memory to disk (e.g., to free up memory for disk/content caching, load other programs, etc). Valid values are between 0 and 100.  On the current release of Ubuntu (Intrepid) this value has a default value of 60 that you can investigate with the following command:

$ cat /proc/sys/vm/swappiness
60

This value is generally fine, but if your physical memory is significantly higher than what you need it’s worth investigating other choices. I personally find a pretty significant performance increase in Rails behavior with a value of 0; essentially telling the kernel to not swap anything out until it becomes absolutely necessary.

The value can be changed either with:

$ sysctl vm.swappiness=0

or

$ echo 0 > /proc/sys/vm/swappiness

A persistent lack of swappiness

If you restart the server, then your change will be lost. If you want the change to remain persistent across restarts you’ll either need to create a script that calls one of these commands or edit /etc/sysctl.conf to specify the swappiness. As always, exercise caution playing with kernel parameters .

Swapping it back in

If you change vm.swappiness and want to force the kernel to swap everything back in you can temporarily disable the swap partition and then immediately re-enable it.

$ swapon -s

Filename                 Type            Size    Used    Priority
/dev/sda2                partition       524280  33948   -2

$ swapoff /dev/sda2
$ swapon /dev/sda2

If the machine doesn’t have enough memory to accommodate the swap-in the command will fail with an error.

Let us know if this helps you out!