Porting from Standard GAE to Managed VM: Part I
by Arthur Messenger

It started out so simple. I found this little module, RandomWords, for generating random words or word lists and I wanted to show how to add this package to the Standard GAE environment (GAE). So I modified the GAE HelloWorld app to say "Hello <random_word>".  This did not work. GAE only allows you to import modules that are written in pure Python and RandomWords compiles a C-shared object as part of its install. Now curious, I found a module, names, written in pure Python that generates random first names, last name, or full names based on the 1990 US Census data (http://www.census.gov/main/www/cen1990.html). This blog post covers what I did to make this work.


HUB.DOCKER.COM and Deleting a Repository
by Arthur Messenger


If you have used the registry at hub.docker.com, you already know that the option to delete an individual tagged entry is not available on the interface.

From what I can tell, there isn’t a way to accomplish this, even in the REST interface. The best I have been able to do is to use the REST DELETE command to delete the repository. This means downloading any images I want to save, deleting the repository, recreating the repository, and uploading these saved images back to the repository.

The rest of this blog is what happened when I used the REST DELETE command to delete the repository.


Interesting Reads
by Arthur Messenger

Interesting Reads


Google Cloud Platform Load Balancer

Google shares software network load balancer design powering GCP networking

Very quick introduction to Google’s Load Balancer showing their sophistication in balancing loads. Examine this with a quick reflection on hardware load balancers in the beginning of web servers, noting how every device used to accomplish these and similar tasks is being virtualized.


Comparing Google Cloud Platform with Amazon Web Services
by Doug Rehnstrom

It’s common when talking about cloud computing services to divide them into Infrastructure as a Service (IaaS), Platform as a Service (PaaS), Software as a Service (SaaS), and Data as a Service (DaaS). So, let’s do that and compare the offerings of Amazon’s AWS and Google’s GCP offerings in each category.


Using Google Cloud Storage as a File Server
by Doug Rehnstrom

When using Google storage, you create “buckets” to put your files in. A bucket is like a server. Buckets contain files and folders. There is no limit on the number of files a bucket can contain. Google storage is designed for high durability, so you don’t have to worry about backing up your files. The server will never be down. The files will be served at the speed of Google’s global network.


Python Virtual Environments on Linux
by Arthur Messenger

Occasionally, I want to test a new module or I have to use a module that I do not want to install in my Python system, or I have to install a version of a module that I will only be using for a project and then not use again. I could use pip -t to install the modules in a different directory and then manipulate sys.path. Then I have to remember how to uninstall the packages and change my sys.path variable again. This is a little error prone and I find myself always having to check to find out what environment I am in.

There is an easier way. The module virtualenv will create a virtual environment that can be activated and deactivated. When it is activated, modules and packages installed are installed in the environment directory. The python executed starts from the environment directory and the sys.path variable is modified to search in the environment directory first.

virtualenv is easy to install through pip, sudo pip install virtualenv.

Here is a very simple example.

  1. Create the virtual environment.
    • Execute: virtualenv Fibonacci
      • See:

Virtual Python1

This will create a new subdirectory under the current directory, called Fibonacci in the example, with several subdirectories and programs inside to set up the environment.

  1. I tend to cd to the environment directory and make a subdirectory called Project-<project_name> just to keep the virtual environment and the code I want to execute under it in the same place.
    1. Execute: cd Fibonacci
    2. Execute: mkdir Project-Fibonacci
    3. Execute: cd Project-Fibonacci
  1. Activate the environment.
    • Execute: source ../bin/activate
      • See:

Virtual Python2

Notice the (Fibonacci) at the head of the prompt string. This is the notification that you are in a virtual Python environment.

Notice to activate you have to source the activate command. The command is in the bin directory inside the environment directory.

  1. Create the environment.
    • This is the installation of the modules and packages needed for this environment.
    • I suggest you create a pip requirements file as a way to easily keep track of the packages added. Recall that in a requirements file you can specify the version of the package you want.
    • For this example, I am not following good engineering practice.
    • Execute: pip install fibonacci
      • See a lot of stuff.
  1. If you can reach the documentation with pydoc the package is most likely installed correctly.
    • Execute: pydoc fibonacci
      • See:

Virtual Python3

Notice the package is stored under the Fibonacci environment directory.

  1. Work the project.
    • Execute: python
    • Execute: import fibonacci
    • Execute: fibo(10)
      • You should see 55.
    • Execute: <Ctrl-D>

  1. Deactivate virtual environment.
    • Execute: deactivate
    • See:

Virtual Python4

Notice the indicator for being in a Python virtual environment is lost.

  1. Clean up.
    • Leave the environment directory intact until you are done with the project.
    • Make sure you save anything in the Project-<project_name> that you need.
    • Remove the environment directory and all the cleanup is done and you do not have to repair or rollback your Python system.




Java and .NET Contrasted
Finding External Components
by David Barkley

Years ago my work took me from Java to .NET (both C# and VB.NET) but recently, as often happens, I found myself living again in the Java world.

I’m sure you know that the .NET and Java platforms and very similar. .NET has a virtual machine, an equivalent to Java Byte Code and built-in class libraries that include much of the functionality of the JCL.

But what are the differences? Over the next few months, I’ll be writing about several of the differences between the two environments—checked vs. unchecked exceptions, getters, setters and properties, primitive data types vs. structs, and perhaps others. In this brief article, I’ll discuss the Java classpath and .NET metadata.

All modern programs reference external libraries. A Java program contains a classpath (set at compile time or via an environmental variable) that points to the location of user-defined classes and packages [Figure 1(a)]. .NET programs, on the other hand, are compiled into assemblies (units of functionality present as either .exe or .dll files) that contain not only intermediate language but also metadata that identifies all external references [Figure 1(b)].

Figure 1

Jave and NET Figure 1

Java classes find required external classes and packages by searching in classpath entries defined at compilation or in an environmental variable. .NET assemblies, on the other hand, contain metadata within the compilation unit that identify external assemblies.


In Figure 1(a) the Java classpath is a pointer to a physical location elsewhere in the environment, but what does .NET metadata point to? In order to understand this, you need to know that there are two kinds of assemblies in .NET—private assemblies and shared assemblies.

Private assemblies are always located in either the same directory (or a subdirectory) of the calling assembly. Shared assemblies can also be located in a directory called the Global Assembly Cache (GAC) (Figure 2).

Figure 2

Java and NET Figure 2

Private assemblies can be located by their calling assemblies, provided that they are in the same directory or in a subdirectory. Strongly named assemblies can also be located in the Global Assembly cache (which is searched first).* The default location of the GAC is %windir%\Microsoft.NET\assembly in .NET 4.0. Earlier versions of .NET placed the GAC at %windir%\assembly.


If you use Visual Studio to compile your projects, your resulting assemblies are private by default. In order to create a shared assembly, you must sign it using the private key of a dual-key encryption system—the public portion of the key is inserted into the metadata of the assembly. (Figure 3).

Figure 3

Java and NET Figure 3

.publickey =  (00 24 00 00 04 80 00 00 94 00 00 00 06 02 00 00  // .$…………..
00 24 00 00 52 53 41 31 00 04 00 00 01 00 01 00  // .$..RSA1……..
E3 C4 A7 AF 7B A6 52 E5 D2 BB 3A 76 30 25 E4 60  // ….{.R…:v0%.`
88 7B 0A 09 3C 75 07 83 85 2E 74 AA 8A 47 94 B8  // .{..<u….t..G..
A5 54 9F DE 98 EB C4 A8 D3 29 DA AB A6 BC 3D 1E  // .T…….)….=.
31 D3 37 27 C4 D6 B8 DF 5C 9D 7B 18 8E 5F 20 8B  // 1.7’….\.{.._ .
9D 93 24 0C FB 0C DD 1B 65 29 4F F5 1F 63 BF 89  // ..$…..e)O..c..
62 67 B2 C4 58 81 33 CC 9D 6A AF 0C 81 29 50 B9  // bg..X.3..j…)P.
92 33 E6 46 2C 9B 0B A1 C0 48 8C DA 8C CB 29 E3  // .3.F,….H….).
BC F2 10 C3 39 13 0C 48 9E 1B AC 2E 3D E9 CA BD ) // ….9..H….=…
.hash algorithm 0x00008004
.ver 1:0:0:0

Shared assemblies are created by signing them with the private key of a dual-key encryption system (called a strong name key file in Visual Studio). A strong-named assembly is a shared assembly and includes the entire public key in its metadata.


When an application sets a reference to an external shared assembly, it includes a token to the public key in its metadata [Figure 4(b)]. If the referenced external assembly is private then, of course, the public key token is missing [Figure 4(a)].

Figure 4

Java and NET Figure 4

Metadata references to shared assemblies include a public key token (b). Metadata references to private assemblies contain only a version number.


There are many more aspects to .NET metadata and private vs. shared assemblies, but this is enough to get you started.

To learn more about how the .NET run finds external references, go to https://msdn.microsoft.com/en-us/library/yx7xezcf(v=vs.110).aspx.

To learn more about the GAC, go to https://msdn.microsoft.com/en-us/library/yf1d93sz(v=vs.110).aspx.


*An application will first look for an external resource compiled as a private assembly in the same directory as its own directory or in a subdirectory with the same name as the assembly. Alternatively, the application’s configuration file can identify another location contained within the directory of the calling assembly. If the external resource is a strongly named assembly, the rules are the same except that the first searched location is the GAC.



Google Cloud Platform
by Doug Rehnstrom

What is Google Cloud Platform?

There are a number of big players in the cloud computing game. Amazon has Amazon Web Services (AWS), Microsoft has Azure, and Google has Google Cloud Platform (GCP).

GCP is built on top of Google’s massively scalable, super-fast and highly available network and infrastructure. Google’s data centers are the most advanced in the world. Google owns their high-speed fiber optic network with data centers and edge caching servers all over the world.

Google offers a wide range of cloud services. These services can be used independently, together, or integrated with services from other providers.


Google BigQuery
by Doug Rehnstrom

What is Big Data?

Organizations today have data from many sources. There are relational and NoSQL databases. There is data in Web logs. There is Google AdWords, Apple iAd data, and Google Analytics data. There’s the CRM system, the call center system, the bug-reporting system, the comments on the blog, the Facebook posts, and the Twitterfeed. You get the idea.

At some point, management will want to analyze this data, and that requires it to be mashed together.