https://martinfowler.com/architecture/

This article on software architecture from Martin Fowler asks the questions “What is architecture?” and “Why does architecture matter?”. The answers are justified through the article which begins with why he does not always like using the term architecture but he accepts it because he believes a good architecture supports programming evolution. Fowler believes that architecture should be defined as “Architecture is about the important stuff. Whatever that is”. Fowler explains that the sole definition of what architecture is has been a debate among computer scientists for a long time but that the definition he puts in this article is the result of a conversation between himself and another professional named Ralph Johnson.

Although Fowler’s definition is his own and very simple it makes sense just like many other computer scientists “definitions” of software architecture. Clearly architecture can be defined in many different ways according to who you ask for a definition but Fowler also states that architecture is important because bad architecture will affect a system in different ways including being harder to modify, having more bugs when updated and these updates/new features being released slower which impacts the customer for which the system was designed. 

Fowler also discusses two different types of architecture which are application and enterprise architecture. Application architecture is based on an “application” which in Fowler’s words is a body of code perceived as a single unit by developers. Enterprise architecture is based on larger amounts of code which work independently from one another in different systems but all used by one company or “enterprise”.

I chose this article as it takes more of a thoughtful approach to the concept of software architecture and in part helps the reader open their understanding of the subject to different “definitions” of the concept which can help when developing software architecture in the future. I feel as though this article would be helpful to anyone in the programming field as it gives you multiple perceptions of architecture, what it is, and why it is important.

This article taught me the general idea of both application and enterprise architecture. I was able to see what Fowler and Johnson viewed architecture as and why they viewed it that way when other computer scientists may define or view architecture in a different light. I learned that architecture itself is very complicated and cannot be defined under one singular definition as it fits many different definitions for many different people in computer science.

From the blog CS@Worcester – Dylan Brown Computer Science by dylanbrowncs and used with permission of the author. All other rights reserved by the author.

Hello all who are reading this, on this post I’ll be discussing a blog that I found in the link just below this paragraph. This particular blog post written by Douglas Minnaar gives a great overview of REST API’s and the knowledge necessary to build them in a structured manner. The blog begins defining the REST acronym, who defined and coined the term, and why one might use the REST style.

https://dev.to/drminnaar/rest-api-guide-14n2

This first part covering the general fundamentals provided a clear and concise picture of how one should structure a REST web structure. In the first part it covers specifically the six architectural constraints of REST. One of the more important constraints I thought was the concept of a Uniform Interface and the ‘principle of generality’ which describes avoiding making a complex interface when simplicity would be far more advantageous across multiple clients.

Another constraint covered was the Layered System. In the blog it is described as more of a restraint for the client, that they should be decoupled in a way that allows the server to work without the client assuming what the server is going to do. It allows the server to pass it through stages like security and other tiers without the client checking back or communicating with it to verify as to not disrupt the process or possibly break security.

Part two shows an HTTP API to go further into the constraints described in part one and how to define the contracts for resources. I felt this part went by a little bit too fast without accurately describing more about what a contract is or what your resources should be but it did show some great points on their naming conventions and how best to rank them, these being techniques I’ll want to apply later should I work further with a REST API. The other sections of part two includes status codes, which I already had learned about in class, and content negotiation. Content Negotiation was very short and only really described as to pay attention to primarily json or xml and throw a 406 code otherwise.

The third part actually gave an example project based off his guide and the Richardson Maturity Model, which in part one is described as a leveled list into how ‘mature’ or as how I read it, how well designed a model is based off how many URI’s a service has and if it implements multiple HTTP methods or status codes. The architecture of the project uses the Onion architecture which I found interesting and almost immediately understood only because of how an onion is structured. The “Ranker” project is mostly an application of the REST architecture and by circumstance is also a movie ranker? It mostly allows you to manage users, movies and ratings but the core of this project is to demonstrate REST and the Richardson Maturity Model and its methodology.

 I felt like this particular blog post gave me some new concepts to think of when working on a REST API as well as some general formatting properties.

From the blog CS@Worcester – A Boolean Not An Or by Julion DeVincentis and used with permission of the author. All other rights reserved by the author.

Christian Shadis               

Docker has been the focus of my Software Construction, Design, & Architecture class for the past couple weeks. In a software development environment, it is paramount that applications can run on different operating systems and in different runtime environments. When applications rely on several dependencies, it becomes cumbersome to install all needed libraries on every machine the application runs on. This is where Docker comes in handy, allowing the programmer to create a self-sufficient container holding all parts of the application, along with all of its dependencies. Anybody can now run an application using its container without having to install any dependencies on their local computer.

Applications, however, are often designed as microservices, each microservice in its own container. A software company may have tens, hundreds, or thousands of different containers that need to be deployed and monitored at once. It is plain to see how this becomes a scaling issue. Container orchestration emerged to address the scalability of containerizing applications. Container orchestrators, like Kubernetes or Docker Swarm, allow the automation of repetitive work related to the deployment and maintenance of containers, such as configuration, scheduling, provisioning, deployment, resource allocation, scale of containers, load balancing, monitoring, and facilitating secure interactions between containers. This is why I chose to read about container orchestration from the article “What is Container Orchestration” from Red Hat.

The article goes into detail on what container orchestration is used for and why it is necessary, along with listing its major functions. It also describes how container orchestration programs like Kubernetes are configured, along with a basic overview of the anatomy and behavior of a Kubernetes cluster.

Using Docker to containerize applications is a pivotal skill for developers to have. In a world where so much computing is moving toward cloud technology, however, it is also important to be able to use Docker Swarm or Kubernetes because a large portion of applications a developer will work on will be deployed on the cloud in some way. In those situations, traditional Docker knowledge will be of little use. Instead, the developer should be able to leverage a Kubernetes cluster or a Docker Swarm to work with large containerized cloud-based applications.

Before reading this article and writing this entry, I had no exposure to container orchestration, though I had wondered about the scalability of the small docker containers we have been working with in class. I learned the basics of the subject, and accrued a list of further references to read more about applying container orchestration in an enterprise setting.

https://www.redhat.com/en/topics/containers/what-is-container-orchestration

From the blog CS@Worcester – Christian Shadis' Blog by ctshadis and used with permission of the author. All other rights reserved by the author.

Application Program Interfaces (API’s) are a set of queries/commands that a user is allowed to use, and are necessary when using microservice architecture. API’s can be considered the messengers between microservices, since they transfer data between services. REST API’s are API that conform to the REST model, which use JavaScript Object Notation (JSON) to transfer data. JSON is a human-readable data format and is used because it is easy to understand. In a blog post “What is an API? A Digestible Definition with API Examples for Ecommerce Owners” Matt Wyatt explains what an API is and why you would want to use it.

API’s use endpoints, which are URL’s that execute a certain method in the API. There are 4 default methods that are standard practice: GET, PUT, POST, and DELETE. GET will retrieve data from the API, PUT will update an existing entry in the API, POST will create a new entry in the API, and DELETE will delete an entry in the API. These methods are just the standard methods, and many others may be created by the developer of the API.

API’s are used in most applications available today. Whenever you do anything online, you are most likely using an API. API’s are so ubiquitous that it would be more difficult to find an application which does not use them. Using Facebook as an example, you would use a login API, a search API, a feed API, friend request API, and possibly many more which are not obvious to the user.

Benefits of using such API’s include increased security, faster response time, and the ability to scale certain features when needed. API’s allow applications to restrict what data they allow in and out, which greatly increases security. This control is paramount when used in an application which contains sensitive information, such as personal details or banking information.

Personally, I believe API’s to be an excellent solution to a wide range of problems. They can easily allow access to data and restrict it at the same time. I see similarities between API’s and Java classes; good Java classes have access methods (getters/setters) which control how the class is used, their attributes are mostly private, and their implementation is kept a secret from the user. A good Java class can be used by someone who has no knowledge of how it works behind the scenes, and the same is true for API’s.

In conclusion, I think API’s are a useful tool that I most certainly will be using in the future. Regardless of my future job, frontend or backend, I will need to either create API’s or use them. They are so universal that avoiding them is almost impossible, although I wouldn’t want to avoid them. They offer a simple solution to a very complex problem, and provide extra benefits along with it.

From the blog CS@Worcester – Ryan Blog by rtrembley and used with permission of the author. All other rights reserved by the author.

Last week we took a look at API Design and discussed the importance of keeping an API as simple as possible. (Learn more about last week’s post here: https://georgechyoghlycs343.wordpress.com/2021/10/28/api-creation-why-simple-is-better/)

This week we will take a look at how this simplicity should be applied not only to APIs but to all programs regardless of purpose or language. As programs become more and more complex it only gets harder to properly keep track of every facet of said program. Having to remember every method and class separately can quickly become a grueling task that eats away at the time, productivity and morale of a software developer.

Within this video the speaker, a professor John Ousterhout, gives their take on software design and what kind of mindset to take when designing your software. One of the major topics of this seminar is the idea of ‘Deep Classes’ which returns to the basic idea of abstraction. Ousterhout bring into focus the issues of ‘Shallow Classes’ which are small classes or methods that provide very little functionality. An example of this is shown below:

private void addNullValueForAttribute(String attribute){
data.put(attribute, null);
}

As Ousterhout states, this is a method that requires a full; understanding of the functionality and does very little to hide information (https://youtu.be/bmSAYlu0NcY?t=918). This method is essentially adding complexity with no benefit which is a net loss in the world of software. With this example in mind, Ousterhout states the biggest mistake people make in software design, “too many, too small, too shallow classes”. They attribute this to what many have been told throughout their career which is to keep methods small. This is problematic because it can actually increase complexity of a program as every class and method adds small amounts of functionality. 

This is especially true in things like the Java Class Library which has many small classes and methods with small functionality. To read a file Java requires you to create three separate classes to give the file input, buffering, and object input. In contrast Ousterhout brings up that UNIX has all of this wrapped into one object that being the UNIX file system which takes care of all other processes such as disk space management in the background.

So why does this matter in the end? The main point to get across is that abstraction is just so important in modern software development. UNIX abstracted its file systems which allows developers to spend little time worrying about file I/O implementation to allow for greater systems to be built. If something is used as often as File I/O, then it is worth it to create an all encompassing class/method for it. As long as classes are well organized there is no reason they cannot be large and have a lot of functionality.

From the blog CS@Worcester – George Chyoghly CS-343 by gchyoghly and used with permission of the author. All other rights reserved by the author.

Microservices architecture is a newer type of architecture that has become popular within recent times. This architecture differs from common architectures like monolith in that microservices “is an approach to developing a single application as a suite of small services, each running their own process and communicating with lightweight mechanisms.” To better understand the microservices architecture, it can be useful to compare it to the classic monolith architecture. In the monolith, there is one single unit, one central server, if you will, that does everything. One of the biggest problems with this design is that if anything needs to be done with the server, like maintenance or something goes wrong and the server goes down, the entire system gets taken down as a result of being one big unit. Microservices fixes this by having many small units/servers running almost independently of each other, so while some may rely on others for some information, if one part goes down for whatever reason, it does not take the entire system down with it.

A good example of how a microservices operates in comparison to monolith is say, for example, we have two main servers in our system, one for the admins, another for the standard users. This way the admins have their server and database, while the users have their server and database. This way, if maintenance has to be done on the admin server, whether it is updating database schema or whatever it may be, this can be done without affecting the standard users, as their server and database are separate from the admin’s server. This is nice as it can limit downtime for the users, so maintenance can be done on the user’s server only when absolutely necessary to limit outages.

Microservices architecture also puts a big focus on failure detection, and trying to automatically restore the service if it goes down. These services are built in such a way that allows early error/failure detection, so that fixes can be put out as fast as possible. Microservices are also built with upgradeability and scalability in mind, so that as a service grows and requires more and more processing power, it is easy to add more servers to add the extra power needed. With monolith, for example, this requires getting a whole new stronger server, rather than just adding another server, since the monolith is ‘all-in-one’, it does not allow the easy upgrading that microservices allows.

Source

I chose the above source from martinfowler.com because we had some readings from that website earlier in the course, so since they had an article on microservices I though what better site to use than them. They also provided a lot of good information on the topic with clear and concise explanations of the topics.

From the blog CS@Worcester – Erockwood Blog by erockwood and used with permission of the author. All other rights reserved by the author.

          Software releases are not always one-and-done affairs; more often than not software that we use is actively being worked on and maintained for an extended period of time well after its initial release. We can expect that software, during its support lifetime, will undergo several types of changes, including implementation of new features and various vulnerability fixes. However, it is important to note that such changes to software are as important to properly document as the technical aspects of the software, such as its use and structure as they were conceived during development. Such documentation of changes is often referred to as “Software Versioning” and it involves the process of applying a version scheme in order to track the changes that have been implemented to a project. While developer teams may develop their own scheme for versioning, some may prefer to use Semantic Versioning (https://semver.org/) as a means of keeping track of changes.

          Semantic Versioning is a versioning scheme that applies a numerical label to a project, which numerical label is separated into three parts (X.Y.Z), which then parts are incremented depending on the type of change that has been implemented. These parts are referred in the documentation as MAJOR.MINOR.PATCH and defined as:

1. MAJOR: version when you make incompatible API changes,
2. MINOR version when you add functionality in a backwards compatible manner, and
3. PATCH version when you make backwards compatible bug fixes.

https://semver.org/

The way semantic versioning works is that, when incrementing the left most part, the progress of the remaining parts is reset to zero, meaning that if a major change is implemented then the minor and patch numbers are reset to zero. Likewise, when a minor change is implemented, the patch number is reset to zero. While this scheme is relatively straightforward in and of itself, the naming convention of the numerical labels (specifically major and minor) may confuse some due to the ambiguity that such names may present. However, there is another naming convention that applies to semantic versioning, which defines the numerical version label as (breaking change).(feature).(fix).

          Though both naming conventions are used, I find the later to be far more straightforward to understand and utilize, as the names can give one a better idea of the importance of a newly implemented update. As I was researching on more resources regarding Semantic Versioning, along with the official documentation, I came across the following archived article on Neighbourhood.ie titled Introduction to SemVer. In this article, Irina goes into further detail regarding semantic versioning by explaining the naming of each component, as well as noting the difference between the two naming conventions.

          Although they go into further detail into semantic release in another article, this article sufficiently covers the fundamentals of semantic versioning. This versioning scheme is not the only way to  software development, it is still an important tool that can help in documenting a project’s history during its support lifetime and outline important changes more clearly and efficiently.

Direct link to the resource referenced in the post: https://neighbourhood.ie/blog/2019/04/30/introduction-to-semver/

Recommended materials/resources reviewed related to semantic versioning:
1) https://www.geeksforgeeks.org/introduction-semantic-versioning/
2) https://devopedia.org/semantic-versioning
3) https://www.wearediagram.com/blog/semantic-versioning-putting-meaning-behind-version-numbers
4) https://developerexperience.io/practices/semantic-versioning
5) https://gomakethings.com/semantic-versioning/
6) https://neighbourhood.ie/blog/2019/04/30/introduction-to-semantic-release/

From the blog CS@Worcester – CompSci Log by sohoda and used with permission of the author. All other rights reserved by the author.

“Interpreter Pattern – Spring Framework Guru”

The above article describes in detail how the Interpreter pattern from the gang of four book works. Before I continue, I should specify that the book, this post and the above article are focused on the Interpreter pattern primarily in terms of object oriented programming, which is not the only way that the general concept may be applied. There is the general idea, which is creating a small, limited language inside of a program, and then there is the gang of four specification, which includes more specific implementation details. For example, it mentions regular expressions, which do not necessarily have to be written in an object oriented way.

I sort of understood what the interpreter pattern was in a broad sense, but not in the specific technical sense one needs to actually make use of it. For example, I didn’t know that it employed the use of another pattern, called the Composite pattern. The Composite pattern is essentially the idea of treating individual objects and compositions of objects uniformly by representing them as a tree. The Interpreter pattern uses the Composite pattern in order to represent an expression to be interpreted. It does this by representing the components of the expression as atomic expressions arranged in a tree called an “abstract syntax tree,” and then expressing the language by recursively processing each node in it. The components of this expression may be either non-terminal or terminal, representing operations with and without operands, respectively (for example, a multiplication versus a constant). The pattern also specifies a “context,” which represents string data that is parsed, and a “client,” which actually parses the expression.

This pattern may be useful in a case where I want to create a smaller, more focused programming language inside of a project for use by less technically oriented people. Specific potential use cases like this include a language for writing plugins in an art program or a simplified language for directing game behavior inside of a game engine like GameMaker’s GML. Other, more common examples include SQL parsing and, as was mentioned previously, regular expressions.

At its core, the Interpreter pattern is essentially about converting instructions from one language to another, much like how a language interpreter interprets ideas from one language to another. “Interpreted” languages such as Python can be seen as an instance of this idea, using an interpreter to convert Python expressions into work by the computer. In the same way, you might also view a C compiler as an interpreter that converts C instructions to machine instructions.

From the blog CS@Worcester – Tom's Blog by Thomas Clifford and used with permission of the author. All other rights reserved by the author.