I just can’t seem to comprehend how Oracle can be so wealthy off of making a database when it’s theoretically easy for a company to store its own information.
In: 0
Spreadsheets are useful for analysis of data that has been filtered and aggregated. Think hundreds or maybe a couple thousand rows that probably contain only the columns you’re interested in. Ex: you have a certain timeframe of sales data and want to calculate or visualize some key metrics, build a few charts, etc. It’s more of a reporting tool that a data storage tool.
A database is a data storage solution. You’re generally using software to interact with the database. Not a DB client, but software that provides features you need and uses the DB as it’s data store (Ex: sales or inventory software, or this Reddit app you’re using to read this). The DB can store much more data (millions of rows or more), organize/structure the data (tables, keys, joins), facilitate efficient retrieval, can be backed up/replicated for high availability and disaster recovery, provide security features and the list goes on.
At the enterprise level you’re looking at many systems like ERP, CRM, manufacturing, HR and more that all use their own DB to store data. Data in the DB would be structured in a way that is useful for the software but probably not so much for a human to interpret. Then you may have some reporting layer of software that allows aggregation of all of this data, to get it into a more human readable form. Then the output of the reporting layer ends up in excel and a bunch of business analysts and managers sit in a meeting looking at it.
Oracle is just one vendor that provides a DB solution (as well as many other software products). There are many other flavors of DBs, some that are free and some that aren’t. Oracle has a big footprint in the enterprise space. A lot of systems are designed to integrate with Oracle as the DB. Large companies have a lot at stake with data storage and trust Oracle to provide support and reliability.
Excel is mostly designed for performing basic mathematical analyses on small amounts of data. And frankly it’s not really very good at that. People do use it for managing finances or performing statistical analyses, but there are lots of specialist software tools for those kinds of tasks that work better.
Database systems are mostly designed for securely storing and retrieving large quantities of data, and for ensuring that it remains consistent. Here are some things that are generally pretty easy with database software but difficult with Excel:
* ensuring that the database won’t be left in an inconsistent state if the system goes down while someone is in the middle of updating records
* retrieving or modifying all data that match certain, possibly complex, criteria
* restricting access so that certain types of data can only be seen by certain users
* logging access so that it’s possible to tell who has seen which pieces of data
* storing extremely large amounts of data (e.g. Excel is limited to about a million rows, a few thousand columns, and a few thousand characters in each cell)
* managing complicated hierarchical data structures, e.g. keeping track of which employees are in charge of which other employees, which buildings they are based in, etc.
Having said that, there are plenty of other options for database software besides Oracle, including numerous free and open-source database systems. Generally, the big advantages of picking a famous commercial application are that it is likely to be well supported and that it will be easy to find staff who are familiar with it.
>when they can just use spreadsheets like Excel
Excel and similar spreadsheet software works fine for storing small datasets but it is not designed or intended to work on the massive volumes of data that large (or frankly, even small) companies need for day-to-day operations. We’re not just talking about the amount of data, but how much new data is being added to it on a daily/hourly/secondly basis (in data science, this is called “velocity”).
>or make their own spreadsheet software?
This is sort of like asking “why would I buy a car from Ford when I could build one in my garage?” The answer is usually some combination of “I don’t have the tools or expertise to do it myself,” “if I spent all my time making (and maintaining) it myself I wouldn’t have time to do anything else,” and “someone else has already designed, built, and tested it, why should I redo all of that work?”
Why are there restaurants if people can just use a microwave? Spreadsheets are good with reasonable amounts of data and not sophisticated data connectors to other systems. When you want to do analysis on gigabytes of data, when you have servers needing to not only report some data but also fetch customer data from all around the world, you have entered the realms of databases. When data needs to be entered by thousands of bots on servers each trying to fetch particular rows and update other rows you need way more that Janette taking phone calls to make changes.
Databases store much, MUCH more information than a spreadsheet reasonably can, and they can manipulate, search, and provide other features around that information more effectively.
You don’t just want to *store* the information. You want to be able to use it for various purposes. You often want guarantees of security or reliability. You want to be able to connect information from different “spreadsheets” (*tables*, in database terminology). You want to be able to do all sorts of performance optimizations. You might need to spread your database across the world for rapid response times, which requires a whole other set of tools. And so on.
Spreadsheets start getting pretty slow if you’re doing anything complex at even a few thousand rows. Databases, on the other hand, can routinely contain *millions* of rows even for relatively low-scale purposes. At scales like that, you need much more sophisticated tools than a spreadsheet can provide.
The most obvious reason is that excel is pretty limited in the amount of rows you can have. It is also a visual representation of the information which isn’t necessary. Databases handle traffic from multiple users better. Databases utilize indexes to make accessing data faster and more efficient. The list goes on.
Excel, Access, etc are fine for small businesses that have a few databases up to a few hundred MB or a GB each, but they don’t operate well at scale. Oracle has built its reputation on being able to house and efficiently run MASSIVE databases… We’re talking in the order of tens or hundreds of millions of cells and many terabytes per instance, with the ability to have hundreds of even thousands of employees working in those databases simultaneously.
Other alternatives just can’t do that.
Spreadsheets have very serious limitations, notably that they cannot be accessed by more than one person at a time.
Doing so can corrupt the file, now imagine having an accounting system built out of hundreds of linked Excel docs that needs to be accessed by a team of 50 accountants. Imagine trying to fix that or reverse engineer it later when there’s problem.
Excel just doesn’t scale that way, which is why you need a database for large datasets and applications like Line of Business Software.
Developing your own in-house software for this is also often a fools errand. It might be cheaper in the short term, but in the long-term factoring in paying/losing developers, developing new features, code efficiency, etc it will cost the company much much more than just buying a canned software platform in the first place.
So why Oracle?
There’s a lot of alternatives these days like PostGreSQL, MS SQL, MySQL, etc etc and many of them are far more cost effective than Oracle.
But a lot of big data applications are legacy meaning that they were purchased ages ago, or are the industry standard, and getting away from running them for a large business just isn’t economical.
The business impact of switching to something else costs a lot more than just continuing with Oracle, which is a fact big software vendors take full advantage of. They know they have their whale customer by the balls so they can basically charge whatever they want.
When there’s a lot of data to be stored and sorted, it can get quite tedious for many companies. Oracle simply makes storing information more efficient and time saving, especially for larger companies.
Excel is not a database program and should not be used as one.
You can force it to do whatever you want but it won’t be good at it. They sell Access if you want a database program
Databases are far far more efficient at storing and more importantly, quickly accessing information
Making your own database and making a database that runs well and isn’t riddled with security issues are two very different things. People pay Amazon/Microsoft/Oracle/Apache/etc to buy a database software that is more efficient and secure than they could make for the same cost
Latest Answers