Databases for companies: The big comparison of relevant databases

Databases are used in every company with its own IT infrastructure. They are required, for example, for merchandise management systems, enterprise resource planning, content management systems and customer relationship management. But databases are also important for online stores. In this article, we would like to explain why you need databases and how they work. We would also like to introduce you to the different types of databases and which ones you could use for your purposes. Stay tuned if your company needs a consistent database to be able to use software like the one mentioned above. If you need support with the use and creation of databases or if you have general questions about software development, we will be happy to assist you as a strong partner.

Reading time 15 min

 

1. Why do you need databases at all? And who needs them?
2. How does a database work? Difference between database and database management system
   1. Types of database models
      1. Hierarchical, network-like, relational, object-oriented, document-oriented database model
      2. Comparison of different databases
         1. MySQL
         2. Mongo DB
         3. Apache Cassandra
         4. Maria DB
         5. Redis
         6. Google Firebase with Realtime Database and Cloud Firestone
         7. Microsoft Azure SQL database
         8. Amazon Web Services (AWS and RDS)
         9. Elasticsearch
         10. Memcached
3. Our conclusion
4. FAQ
   1. Where are databases needed?
   2. What advantages do databases offer?
   3. When is a database worthwhile?
   4. What does SQL and NoSQL mean?
   5. What does sharding mean?
   6. What is ArangoDB?
   7. What is a graph database?

 

Why do we need databases at all? And who needs them?

Databases can be used to manage electronic records and data. The data is collected in the database and logically structured. Assignments and access rights can then be defined in more detail using a database management system. In order to actually use a database, a suitable database application must be used. This can then not only be used to open the database, but also to read and edit it.

So much for the very technical definition of a database. Put simply, a database allows you to manage databases (within your company, for example) and retrieve important information in a timely and bundled manner. It is particularly suitable for displaying content clearly so that it can be reused. You also have the option of assigning rights as to which employee or which system can read which data. This helps your company with customer management and processing, for example. A database should be distinguished from a collection of files in a folder on your computer. The term database is often used, although it is not technically a database.

In fact, every company with an IT landscape needs a database. Because if software is used or data is collected and queried, a database system must be used. Web applications such as CRM, ERP and CMS, for example, also work on the basis of a database. This allows relevant information to be stored and displayed to a user. Certain user paradigms are important for an ERP web application. For example, transactions must be processed quickly and without delay. For these use cases, it is important that the database system can be used with an OLTP. OLTP stands for Online Transaction Processing or real-time transaction processing.

How does a database work? Difference between database and database management system

A database can only function with the help of a DBMS, i.e. a database management system. It manages data, processes data volumes and also serves as data storage. A DBMS is an essential part of a database. You cannot use the database without a database management system. Nowadays, almost all software is based on a database system.

The main difference between a database and a database management system is this:

The database stores the data types and is also known as the database. The database management system is the software used to manage the database. Without it, it would not be possible to retrieve the information stored in the data records. It determines the access rights and the structure of the data. As a rule, a separate database language and a database model are also used for administration. The database model is responsible for the architecture of the database system. One type of database management system is the Relational Database Management System, or RDBMS for short. Well-known RDBMSs include Ingres, Oracle and Sybase.

Types of database models

A database model is a collection of concepts that can be used to describe the structure of a database. It also contains the structure of a database, including the relationships between different data elements. In addition, it usually contains information about how data is stored, retrieved and updated.

The most common database types are the relational model, the network database and the hierarchical model. In a relational model, data is organized in tables and fields, while in a network model, data is organized in nodes and arcs. In a hierarchical model, the data is organized in a tree structure. Various query languages or data query languages can be used to search for information in the databases. A query language is therefore a database language of the respective database.

Each of these database types has its own advantages and disadvantages. The relational model is the most popular model because it is easy to understand and use. The network model is better suited to processing large data types and quantities, while the hierarchical model is better suited to organizing data types in a specific structure. There is also the object-oriented and the document-oriented database model. We would like to explain these database types in more detail below:

Hierarchical database model

In this type of model, the data is arranged in a tree-like structure, with each data element having a single "parent" and zero or more "children". We would like to use the following example to explain the tree-like structure:

Let's say we have a database of animals. Each animal in our database has a species (e.g. dog, cat, bird, etc.), and each species has several different subcategories (e.g. dog breeds, cat breeds, bird species, etc.). In a hierarchical model, this data would be organized as follows:

Animals

- species

 

-- Dog

--- Golden Retriever

--- Labrador Retriever

--- German Shepherd Dog

 

-- Cat

---Siamese

--- Bengali

 

-- bird

--- Parakeet

--- Finch

 

Each part of the data in our example - animals, species, subcategories - is called a "node". In the hierarchical model, the nodes are arranged in a specific order, that is, each node has exactly one parent node (except for the very first node), and each node can have zero or more child nodes. This relationship between the nodes is called the "parent-child relationship".

The hierarchical model is thus very efficient when it comes to storing data that can be classified into subsets. This is because it allows you to find all animals that fall under a certain category (e.g. all dogs, all cats, all birds) without having to sift through unrelated entries. For example, if you want to add a new dog breed, you simply create a new node and add it to the right place in your hierarchy.

One drawback of the model, however, is that it is difficult to represent the relationship between data that cannot be divided into subsets. In our example, it would be difficult to mark animals as siblings. In addition, retrieving information from a hierarchical model can be quite slow. Since each piece of data can only have one parent node, we often have to traverse the hierarchy multiple times to find the information we are looking for. For these reasons, the hierarchical model is generally not used in modern databases.

Network-type database model

In a networked database model, the data is organized in a series of linked nodes. Each node represents an entity, e.g. a piece of information or an object. The links between the nodes represent relationships. An important aspect of the networked database model is that it allows relationships between multiple people. This means that a single node can be linked to several other nodes and vice versa. In contrast, other models such as the hierarchical model only allow one-to-many relationships, as we have already explained above.

The main advantages of the network database include flexibility, efficiency and scalability. The main disadvantage is that querying data in this model can be difficult. A network-type data model is suitable for a system that needs to store a large amount of data and allow users to access the data from multiple locations.

Relational database model

Have you ever noticed how easy it is to search and retrieve information on websites? Chances are, the website uses a relational database model to store and organize its data.

So what exactly is a relational database model? In short, it's a way of organizing data into tables that are linked together by common attributes. These links make it easy to retrieve and update information. Each table contains rows and columns, with each column representing a specific attribute and each row containing a specific record. A unique primary key is used to uniquely identify a data record. A distinction can be made between a unique, composite and artificial primary key. For example, in a customer database, each row would represent a single customer, and the columns could contain things like name, address, phone number, etc. The relational aspect comes into play. The relational aspect comes into play because there can also be links between these tables - for example, a separate "orders" table linked to the customer table via a customer ID. This makes it easier to call up all orders for a particular customer without having to manually search through multiple records. Overall, relational databases provide an efficient and organized way to manage large amounts of data. A relational database model is suitable for both large and small companies.

Object-oriented database model

In contrast to conventional models, object-oriented databases store data as objects or individual units with their own attributes and relationships. This allows for greater flexibility and easier manipulation of the data. In addition, object-oriented databases usually offer support for object-oriented programming languages such as Java, making them a popular choice for software development.

The object-oriented data model is best suited for storing complex interconnected data. In this model, data is organized into objects that contain both attributes and behaviors. For example, a person object can have attributes such as name and age and behaviors such as "retrieve address" or "change phone number". These objects can then be linked to other objects, creating a network of interconnected information. This makes the object-oriented model perfect for managing data in areas where there are many relationships between entities, such as finance or healthcare. It also allows flexibility in adding new attributes and behaviors without having to restructure the entire database. So if you're dealing with complicated, interrelated data, the object-oriented database model may be the right choice.

Document-oriented database model

A document-oriented database model differs from other databases in its structure and design. Instead of organizing data into rows and columns, information is stored as "documents" that contain all the necessary information about a particular topic or entity. This makes it well suited for managing unstructured or rapidly changing data, such as blog posts or user profiles. It also allows for faster retrieval of related information as the documents can be easily linked together. Overall, a document-oriented model can provide an efficient and flexible solution for managing dynamic data.

Comparison of different databases

Now you are familiar with the different database models and know roughly who or what they are suitable for. But now you probably want to know which databases really exist, what features they have and so on... For this reason, we have taken a closer look at the most common databases and analyzed them in more detail. We want to show you which advantages and disadvantages the databases have and which details can help you decide which database is right for you. We will take a closer look at the following aspects for each database:

• Type of database model: we'll look at which type of database model the database in question fits in with the database models presented previously.
• Licensing: What about licensing? In which models can the database be "purchased"?
• Database speed: Of course, the performance of your systems is incredibly important for your business. That's why we take a closer look at the speed of the databases.
• Suitability for large amounts of data: Do you want to be able to store, retrieve and process any amount of data? We look at which database is suitable for your requirements.
• Simple administration and administration tools : Your database should also be easy to manage. We check how the administration works and whether tools are available to you.
• Cloud or on-premise: A cloud database is located on servers and storage and can be easily accessed via the internet. On-premise solutions, on the other hand, are located on your company's local network. Depending on the intended use, one is more suitable than the other.
• Compatibility with operating systems: Which operating systems is the database compatible with? Or is it even platform-independent?
• Support: Does the company offer support in the form of live chats, e-mail support, a telephone hotline or a FAQ in the event of problems with the database?
• Open source: Is the database open source? If so, you can access the code.
• Paid features: Are there additional features or functions for your chosen database version that are also subject to a fee?

MySQL

MySQL was developed in its current form by the company Oracle. It is a database that is particularly suitable for dynamic websites. For example, it is used for all WordPress sites, but the database can also be used for online stores with Shopware.

MySQL advantages	MySQL disadvantages
There is an open source version The database is reliable (high availability) Fast database query High compatibility with other systems, programming languages etc. High security	High costs for licensing the enterprise version A crash could permanently damage the system catalog

• Type of database model: Relational database management system (also use of the client server model)
• Licensing: Open source free and expensive commercial version
• Database speed: Generally high speed, fast data retrieval
• Suitability for large amounts of data: Yes, but performance deteriorates as the amount of data increases
• Simple administration and administration tools: MySQL is easy to learn how to use
• Cloud or on-premise : Both
• Compatibility with operating systems: Suitable for various operating systems
• Support : Oracle offers, for example, telephone support and 24/7 live chat for customers
• Open source: Available as open source software and commercial enterprise version
• Paid features: Yes

 

 

MongoDB

MongoDB is a document-oriented database system that was released in 2009. It is written in the C++ programming language and runs on the Microsoft Windows, Linux, macOS and OpenBSD operating systems.

MongoDB offers different versions. There is the free Mongo DB Community Server for Windows, OS X and Linux, the Mongo DB Enterprise Server for commercial use and Mongo DB Atlas, an on-demand cloud database for GCP, Azure and AWS. Overall, MongoDB has three different pricing tiers: shared for $0 per month, dedicated for $57 per month and serverless starting at $0.30 per million reads.

MongoDB Advantages	MongoDB disadvantages
Flexible and adaptable when it comes to unknown data structures Good for unstructured data Changes can be made easily Good scaling Storage of different file types is possible Tech stack is not affected	Complex transactions Memory consumption is quite high There is no integrated analysis of memory consumption No stored procedures, i.e. no implementation of your business logic at database level

• Type of database model: NoSQL (i.e. not relational)
• Licensing: Free and paid versions
• Speed of the database: High speed
• Suitability for large amounts of data: Yes
• Simple administration and administration tools: Administration is simple
• Cloud or on-premise: Both
• Compatibility with operating systems: Compatible with many operating systems
• Support : For Mongo DB Community there is the Mongo DB Community Forum, for Mongo DB Atlas a support ticket can be opened in the account, for MongoDB Enterprise a ticket can be opened via the Mongo DB Support Portal
• Open Source: The community version is open source.
• Paid features: Yes

 

 

Apache Cassandra

The Apache Cassandra database was launched on the market in July 2008 by the Apache Software Foundation. Cassandra is a NoSQL database system that is particularly suitable as a database for companies. This is because the simple database management system has a high level of reliability and is easily scalable.

Apache Cassandra advantages	Apache Cassandra disadvantages
Speed advantage over other database programs Support for MapReduc High scalability of the database program (horizontal) High reliability Cross-platform compatibility with many operating systems High fault tolerance	Performance problems can occur if the data model is poorly designed Due to the architecture distribution, replicas may be inconsistent

• Type of database model: NoSQL
• Licensing: Open source Apache license version 2
• Speed of the database: High speed
• Suitability for large amounts of data: Yes
• Simple administration and administration tools: Yes
• Cloud or on-premise : Both
• Compatibility with operating systems: Is given
• Support : Paid support from DataStax
• Open source: Available
• Paid features: No, not available

 

 

MariaDB

MariaDB was developed as a database management system from MySQL. The enterprise version of MariaDB is primarily a database for companies. It is suitable for business-critical applications due to its high performance. The MariaDB database was released in October 2009 and works in combination with the Microsoft Windows, OpenBSD, Solaris, macOS and Linux operating systems.

MariaDB advantages	MariaDB disadvantages
Fast backups High performance MariaDB is suitable for business-critical applications	After switching to MariaDB, you can no longer switch back to MySQl

• Type of database model: Relational database model
• Licensing: Maria DB Enterprise Version and Maria DB Community Server
• Speed of the database: High
• Suitability for large amounts of data: Yes
• Simple administration and administration tools: Yes
• Cloud or on-premise: Both
• Compatibility with operating systems: Yes
• Support: Contact form or as a customer MariaDB support
• Open source: Open source license for enterprise and community version
• Paid features: No

 

 

Redis

Redis is an in-memory database, i.e. a database that uses the main memory of a computer to store data. It is therefore also a NoSQL database and was released in April 2009. Redis is particularly suitable for simple data structures and is also used as a cache for websites and more.

Redis advantages	Redis disadvantages
Redis is faster than relational databases Vertical and horizontal scaling possible	Not suitable for complex data Redis requires a lot of memory, which is associated with costs Risk of data being lost when the computer is off, as data is not persistent

• Type of database model: NoSQL
• Licensing: Various licensing options from open source to Redis Enterprise
• Speed of the database: High
• Suitability for large amounts of data: No
• Simple administration and administration tools : Yes
• Cloud or on-premise: Both
• Compatibility with operating systems: Yes
• Support: Commercial support for the enterprise version or community support from the community itself
• Open source : Yes
• Paid features: No

 

 

Google Firebase: Realtime Database and Cloud Firestore

Cloud Firestore advantages	Cloud Firestore disadvantages
iOS, Android and web SDKs with offline data access High level of user-friendliness Automatic scaling Real-time updates High reliability Can be used with programming languages such as Java, Unity, Node.js, Go and C++ SDKs	Targeted and optimized for apps, poor support for other applications

Features of the Cloud Firestore:

• Type of database model: NoSQL
• Licensing: A free version, but otherwise charged according to database storage, bandwidth and number of transactions
• Speed of the database: High
• Suitability for large amounts of data: Suitable for complex data
• Simple administration and administration tools : Yes
• Cloud or on-premise: Cloud
• Compatibility with operating systems: Cross-platform
• Support : Google Support
• Open source: No
• Paid features: Yes, other Firebase products

 

Google Firebase advantages	Google Firebase disadvantages
Real-time synchronization of data between different users Asynchronous synchronization for offline use	Data can only be saved as a document tree, data organization is difficult

Features of the real-time database:

• Type of database model: NoSQL
• Licensing: Free model available, otherwise different tariffs
• Speed of the database: High
• Suitability for large amounts of data: Only suitable for simple data
• Simple administration and administration tools : Yes
• Cloud or on-premise: Cloud
• Compatibility with operating systems: Platform-independent
• Support : Google Support
• Open source : No
• Paid features: Yes, other Firebase products

 

 

Microsoft Azure SQL database

With Microsoft Azure, Microsoft offers a cloud computing platform that was created for software developers. Microsoft Azure offers databases that can be used in the form of a cloud model.

Microsoft Azure SQL database advantages	Microsoft Azure SQL database disadvantages
High level of data protection and strong focus on security High availability and scalability Attractive cost model For server hosting, the country for your hosting is freely selectable	Specialist knowledge is required to use the database efficiently

• Type of database model: SQL
• Licensing: Three levels: Basic, Standard and Premium
• Speed of the database: High
• Suitability for large amounts of data: Yes
• Simple administration and administration tools : Yes
• Cloud or on-premise: Cloud
• Compatibility with operating systems: Platform-independent
• Support: Various support plans
• Open source: No
• Paid features: Yes

 

Amazon Web Services (AWS)

Amazon Web Services, or AWS for short, is actually a cloud computing provider that offers various services. These are also used by well-known companies such as Netflix and Reddit. In addition to servers, storage and networks, AWS also offers a wide range of databases. These are the virtual databases SimpleDB or RDS (Relational Database Service). SimpleDB can be used to store simple, i.e. non-relational information. However, the database is still in the beta version. For this reason, in the following details we will only deal with RDS, which is already officially supported and can be used. It is a service from Amazon that makes it easier to set up, operate, manage and scale databases in a cloud. It is therefore not a stand-alone database. However, the service includes various databases that can be used as part of the pricing model. Among other things, Amazon also offers its own relational database service Amazon Aurora.

Due to the variety of services offered, it is worth considering RDS and therefore AWS.

Amazon Web Services advantages	Amazon Web Services disadvantages
No need for own server and IT infrastructure due to cloud model Six different database engines can be selected (Amazon Aurora, MySQL, PostgreSQL, MariaDB, Oracle and Microsoft SQL Server) Flexibly scalable Local databases can be migrated using RDS Database software is always up to date	The services of AWS and RDS are comparatively expensive. For large companies, however, it is worth the cost, as the service is really powerful and the reliability and scalability are excellent.

• Type of database model: Relational
• Licensing : Billing model based on selectable storage and computing power, in the SaaS service model from Amazon Web Services
• Speed of the database: High
• Suitability for large amounts of data: Yes
• Simple administration and administration tools : Yes
• Cloud or on-premise : Cloud
• Compatibility with operating systems: Yes
• Support : Support via AWS
• Open source: No
• Paid features: Yes

 

Elasticsearch

Elasticsearch is not a classic database, as it serves a more specific purpose. Elasticsearch is particularly suitable for document and text searches. This is why it is also referred to as a search engine that works on the basis of Lucene. Apache Lucene is in turn a program library for full-text searches. Elasticsearch was launched by Elastic in 2010 and is platform-independent.

Elasticsearch advantages	Elasticsearch disadvantages
Fast search results and queries for text searches Well suited for online stores, for example in combination with Shopware Simple operation Supports programming languages such as PHP, Java, Python, Ruby and JavaScript (Node.js)	Installation and maintenance costs should not be underestimated, but are still manageable Elasticsearch is not available in all store systems

• Type of database model: NoSQL
• Licensing: Free Server Side Public License, for cloud offers the Elastic License for a fee
• Speed of the database: High speed
• Suitability for large amounts of data: Yes, suitable for large amounts of data
• Simple administration and administration tools: Yes
• Cloud or on-premise: Both
• Compatibility with operating systems: Platform-independent
• Support: Community page and contact form at the Elastic company
• Open source: No
• Paid features : Available, if desired

 

Memcached

Memcached was developed in 2003 by Brad Fitzpatrick and is a widely used open source software project that acts as a distributed in-memory cache. It was developed to improve the performance of web applications by caching data in a fast and efficient memory.

Advantages

Disadvantages

Fast access to data: Memcached stores data in main memory, resulting in very fast read and write accesses. Scalability: Memcached is able to store and manage large amounts of data. Database load reduction: By caching data in the memcached cache, frequently used data can be retrieved directly from the cache without having to access the database. Easy integration: Memcached offers APIs for various programming languages, which facilitates integration into existing applications. Flexible data modeling : Memcached uses the key-value data model, which is simple and flexible. Multi-threaded architecture: Memcached is optimized for efficient use of multi-core processors.

Limited functionality : Memcached is limited to the core functionality of a cache, namely the storage and retrieval of key-value data No security mechanisms: Memcached offers no integrated security mechanisms. There is no standard authentication or encryption of data traffic between the clients and the server. Memory requirements: Memcached stores data in the main memory, which leads to increased memory requirements. No automatic data invalidation: Memcached offers no integrated mechanisms for automatic data invalidation. This means that if data changes in the data source, the data stored in the cache is not automatically updated.

• Type of database model: distributed in-memory cache
• Licensing: open source, free of charge
• Database speed: high speed, fast performance
• Suitability for large amounts of data: Yes
• Simple administration and administration tools: Yes
• Cloud or on-premise: Both
• Compatibility with operating systems: Yes
• Support: Community support, support from developers
• Open source: Yes
• Paid features: No paid features available

 

 

Our conclusion

There is now a huge selection of databases on the market that are suitable for various purposes. In our comparison, we have already compared some databases such as MariaDB and Redis based on a few specifications, but by no means all of them! For example, there are also SQLite, CouchDB, Ninox and PostgreSQL. Perhaps we could give you a good market overview of the databases on offer. However, if you still don't know which database is the best choice for you, please feel free to contact us. Because every IT project is different and has different requirements. It is therefore impossible to say which database is the best. However, we are happy to recommend the best database for your purposes and offer you help with your IT and software projects .