Data Structures

 Structuring Data

Here’s the scenario…

…You’re learning something new. And while the instructor is yielding the information to you and your class group, like the desultory methods of any other unorganized fellow, he jumps from subject to subject. So within a couple of hours you learn a few minutes of mathematics, yet still a few of social studies, and still again some other rhetorical learning subject, with none in such a way that you might utilize to make sense of.

And while I hate to break the good news to you, this is precisely how most of what we learn is constructed, making it especially difficult to grasp, hold on to or even connect the dots so as to form larger conceptualizations from the information we receive…pretty unorganized wouldn’t ya say?...yeah I must say I agree.

Not to worry though, our exceptionally smart machine developers have devised methods that help us better organize information so that the data we store in our software applications give us efficiency aim at grouping and retrieving data so that we mightn’t make the same mistake as our poor old desultory teaching instructor.  

This is what the use of data structures in our computer programs help us do.

How so Omnitekk?... 

Hmmm… let’s us have a look.

Just as we use specific methods of organizing the information we receive, to efficiently process and retrieve at some interval, it is always good programming practice to store relative clusters of data within groups of structural data sets, in our effort to improve data retrieval efficiency within our software applications.

Further, the use of data structures allows developers the added benefit of utilizing computer memory and addressing in such a way that the seasoned developer can make the best use of minimal storage assets within the computing ecosystem.

Information is stored in most machines using addressing schemes that make efficient use and retrieval of the groups of information closest to it. 

For instance, isn’t it far easier for you to retrieve an item that just happens to be closest to you than one farther away?

This is the same logic our computer folks used to design the addressing schematics within most computing machine environs.

And while fallacious to say that all data structures are stored in a sequential manor, when we programmatically group data, our programs do have the means of accessing relative information, such as records in a file, in such a way that allows machine processing methods to be implemented with higher efficiency.

But wait…it gets better!...

…we haven’t even covered all the nooks and crannies of the “Data Structure”…So shall we?

The Casting Call

When it’s cold outside, most none of us would wear shorts or t-shirts – except maybe those guys who polar bear dip in the arctic, but we’re not them…now are we? 

Any who, when it comes to the types of data structures or data organization methods we use to store, process and retrieve the data within our software applications, the data structure types we choose should largely depend on quite a few factors…such as

How The Data Will Be Used 

A few things to consider here is whether your software application is going to simply store, and retrieve data or shall it be some culmination of data processing types? Likewise, will the data be sorted or in unsorted form, or will it have a one-to-one data correlation or a one to many or even a many to many data relation? 

Data Size 

Some data structures yield far greater efficiency with smaller datasets than others, especially when it comes to data retrieval. You might ask yourself whether your application is designed for miniscule data inputs or whether you’re designing an enterprise level application. 

knowing the expected data size design specifications should help you choose the most efficient data structure to use for your project.

Data Structure Order

Some data is best used in first-in-first-out (FIFO) order, such as in a data queues, which we typically see in our lines at most stores. The first one to arrive is usually the first one to be attended to – unless someone cuts the line of course. There are also information storage processing types that utilize last-in-first-out (LIFO) processing method, such as stacked data or even the clothes you layer on. The top layers(last ones placed) are usually the nearest accessible(first one to come off)…like your coat for instance.  There’s also data priority structures, such that information is processed by its most critical use or priority.

Data Complexity

Some data can be easily grouped for storage, processing and retrieval using simple methods of structural data formations, while other data design methods are best constructed using a complex data structure to best represent the nature of the data solution.

Knowing whether your design goals call for a simple implementation, such as array data structures or something of higher complexity, such as a graph or map data structure to best represent your programming application can mean the difference in designing an efficient programming solution vs. one that’s inefficient

Data Type 

Some data structures work best with string data, such as string-matching algorithms, while others for other types of data, such as numerical data.

The Structures

There’s an array of data structures we might use to group our data for storage, processing and retrieval…ARRAYS being one of them…no really.

So let us have a look at them shall we?

Arrays

Array Data Structures are linearly organized or daisy-chain like data structures that are grouped in sequential addressing fashion where all data is adjacent to neighboring data.

Likewise, with few exception, array sizes are usually predefined prior to structure implementation, making them unfavorable if data is expected to grow or shrink exponentially. 

Linked Lists 

We might think of Linked List Data Structures as chain-like data grouping programming methods that follow a sequential forward or backward data retrieval scheme, where data is stored and retrieved usually by traversing or traveling the list using the neighboring data nodes within the list as counters into resident list data. 

And while linked list insertion and retrieval methods can certainly prove unorthodox, in retrieval method, especially for the crafty or artistic developer, they do typically follow specific implementation methods.   

Stacks

We’re stacking crates one on top or above the other. As we remove the stack of crates, so as to retrieve the contents, the last one placed, is usually the first one retrieved. This is the way Stack Data Structures work. They are usually last-in-first-out (LIFO) data structures, meaning that the data placed using this type of structure is retrieved in reverse the order they were placed. 

Queues

We’re waiting in a long line, a little anxious of course, but who isn’t. we just want to pay for our things and spend what’s left of our day salvaging through our purchases, marveling at what we’ve bought. Those in the line are usually processed in the order of arrival. And this my friend is the Queue Data Structure.  

With queues, the data retrieved is usually that which was first placed, as they are  first-in-first-out(FIFO) data structure types.

Trees

Now here my friends is where our data structure journey can get a little tricky.

Tree Data Structures are structures whose data connections can represent one-to-one and likewise one-to-many clustered data representations, making data traveling or traversal within the tree structure quite intricate.

Tree data structures follow the notion of a predecessor, successor nomenclature or naming convention whereby successor nodes are linked to specific predecessor nodes, which are linked to new groups of successor predecessor node groupings and so forth, all stemming from a single predecessor.

The basic concept of trees is in their clustered or branch-like nature, just as the branch, leaf, and bark structural leanings of actual trees.

Priority Queues And Heaps

Priority Queues function much like ordinary queues, with the exception that they don’t necessarily follow the stringent ordering of our dear unprioritized FIFO data structures. Instead, with priority queues, data is usually process in some contrived order of importance. So the data entered into the queue isn’t necessarily the data that will be processed in fashion as it was entered, or in other words, priority queues don’t necessarily follow a sequential pattern of data access or retrieval.

In the other cart on our train session here, Heap Data Structures can be thought of as trees that typically follow a sequential hierarchal priority data retrieval method, such that the predecessor is either greater than or less than the successor value nodes.

So like the priority queue, the storage and traversal method of accessing data in the queue is contingent on the neighboring node cluster value in that it’s value is connected to the way data is placed onto the tree structure by order of numerical value. 

The tree is either growing in value or shrinking, meaning the tree data values are increasing or decreasing as we traverse the heap left and right.

Disjoint Sets

As the name implies Disjoint Set Data Structures are open clustered tree structures that typically don’t follow the normal tree design methods of implementation. They might be thought of as incongruent tree data structures, such that the “clustered” tree connections or data nodes might be formed from numerous trees joined together to construct an accessible data relation or set of data items. 

Graph Algorithms

If you haven’t already figured it out yet from our provisional expose, data structure design and implementation methods increase in complexity from simple structures to ones that are quite complex, just as many real-world examples, such as a data representation of houses on a single street, requiring only a linear or simple data construct to represent versus the representation of a whole city, state, region or continent locality information. 

Such is the notion with Graph Algorithm Data Structures. 

Graph algorithms are structures similar to trees, except instead of representing a one-to-one or one-to-many data relation of connected predecessor successor pairings, graph data structures can represent many-to-many data representations, meaning there isn’t necessarily a single predecessor successor dynamic.

A group of successors nodes can have numerous predecessors, or we might think of these structures as a poly or matrices data representation instead of hierarchal ones.

Hash Tables

Hash Table Data Structures, like graph data structures, can depict many-to-many data representations. 

Unlike graph representations though, with hashing, the data represented is usually indexed using a key-pair data retrieval scheme that allows a single key to index a record of data values. 

With hash tables, the many-to-many relationship structure uses a few crafty tricks of the trade to retrieve numerous values using one key indexer.

Deconstructing

And that’s it folks, we have reached the end of our data structure journey...Data Structures Served...

I should probably note that there are a few lesser used, really complex data structures, such as Data Maps and Data Dictionary structures, but they aren’t much different than the ones presented here.

In fact, they are best described as the culmination of those listed.

I hope you enjoyed our walk down Data Structure Lane.

And until our next IT adventure my friends, OMNITEKK says be well.