Database types and a countdown of the best databases of 2020
This post is also available in : Spanish
Hybrids, in-memory and as a service: find out which are the best databases of 2020
We find ourselves at a very interesting time for the database sector, and there are many types of databases on the market as of now. On one side we have relational databases, which right now have got a stronger establishment. On the other side we find NoSQL databases, which have really been on the market for over a decade. Along with these two types of databases we can find hybrid databases (SQL & NoSQL), in-memory databases and databases as a service.
If you wish to add information don’t hesitate on dropping a comment with your experiences so we can all contribute to enriching this article.
Databases may well be one of the most important parts on any installation. This is due to its use as the main data repository. This data has two characteristics that make it vital for any business activity:
- Without them our business probably couldn’t function.
- It’s data that has to be accessible in real time for all customers.
For these reasons, database monitoring is priority on any installation.
In any project we recommend doing a good and careful selection process when it comes to deciding which database to use, because our success will depend greatly on it. Not only should we think about the present time we’re in, but also we should evaluate our database thinking about where we would like to be.
Frequently asked questions when deciding on a database:
- How many clients do I want to frequently serve?
- How much data will I have to manage?
- Will I need to apply “batch” work to access the database?
- What response time will I need to give to my clients?
- How am I going to scale my database when the number of clients and transactions starts raising?
- How am I going to monitor my database in order to achieve the least unavailability time possible?
- I need a relational or NoSQL database
- How will my database behave in case it goes down? How will it respond when it encounters problems?
Up next we’re going to go ahead and list the most important databases to consider during 2020, along with their main features and specifications.
Best commercial databases
Nowadays the market is still owned by DB2, SQL Server, Oracle and IBM. On Mainframe or Unix/Linux systems, Oracle and DB2 are the absolute leaders, while among all operating systems Windows SQL Server is usually your best choice..
A database that can run on most any operating system. From Oracle we highlight the abundance of experienced professional profiles that have come along with this technology, apart from the large amount of tools available to manage and monitor it.
Oracle benchmark: http://www.oracle.com/us/solutions/performance-scalability/index.html
It’s usually the second most used database on Unix/Linux environments, after Oracle. It’s clearly a disputable winner when it comes to Mainframe. There are professional profiles specialized in DB2 but not as many as for Oracle. On the other hand, professional profiles for the DB2 mainframe do not necessarily need to know how to use DB2 on Linux/Unix.
DB2 benchmark: http://www-01.ibm.com/software/data/db2/performance.html
Microsoft SQL Server
Database with exclusive compatibility for Windows systems. There are a lot of expert profiles for SQL Server and it’s not a difficult expertise to acquire. Its integration with Microsoft Azure has improved its flexibility and performance quite a lot.
Stands out from the rest of databases because of its storage capacity and data analysis. It’s usually applied with Big Data installations.
Even though it’s not in its prime like a decade ago, it still excels in scalability and performance.
Best free distribution relational databases
Now we’ll get into talking about the best open relational databases. Its benefits are: community, the chance to see and even modify the source code as we see fit, all along with the fact that payment is not mandatory.
The most important databases are MySQL, Maria DB and PostgreSQL.
Document oriented NoSQL
Here documents that support various formats can be stored (JSON, XML). You can change schemes without interrumping the database and the developers can insert indexed databases with ease using the database’s motor.
This is probably the most popular database. It has the possibility of working with both structured or unstructured data. It stands out for its great scaling capacity and its performance. It’s usually the one with the most amount of experienced professional profiles.
They allow working in pairs formed by keys and values, as well as accesing different parts of the stored data.
Mongo DB doesn’t have support atomicity in operations and guarantees event integrity. Changes will be replicated along all nodes but it’s not guaranteed that all nodes have the same piece of data at the same time.
Open database built on the Apache project. Also doesn’t guarantee 100% data integrity. It excels for its great administration console which can access a lot of data in very simple ways.
Mark Logic Server
This database stands out from the rest because it allows data integrity and is compatible with XML, JSON and RDF.
Supported systems: Windows, Solaris, Red Hat, Suse, CentOS, Amazon Linux and Mac OS.
Other databases that can are worth mentioning are RavenDB, Apache Jena and Pivotal GemFire.
No SQL databases oriented to password protected values
Ideal when you have to access data with a key or password. The difference with these databases is rooted in the possibility to store data without a predefined scheme. They’re the simplest databases to use and are usually quite efficient when it comes to reading and writing. Data is usually stored in complex structures like BLOB.
Stands out for its capacity of being a database based on password-value pairs, document storage and being prepared for search queries.
Column Oriented No SQL’s
These are databases that allow mapping passkeys and values, grouping them into structures. Used in environments where there is little writing and the need to access various multi-lined columns is present.
Very useful when processing and analysing events, managing contents and analysing data.
Database created by Facebook which is now freely distributed. We can recommend this database to manage insanely massive amounts of data.
The enterprise version to Cassandra is Datastax Enterprise.
Supported data formats: ASCII, bigint, BLOB, Boolean, counter, decimal, double, float, int, text, timestamp, UUID, VARCHAR and varint.
Designed to withstand large quantity of reading and writing accesses to large amounts of data, all in real time. One of its advantages is that it rubs on Hadoop and the Hadoop filesystem.
Databases oriented to Non-SQL graphs
Using the graph theory to link data from the database. Every element points to its adyacent element. These databases are recommended if your data is very related among itself like what happens in social networking, fraude detection, real time recommendations, etc. Here, the database must be standardized where each structure will have a column and each relation, two.
Supports data integrity and high availability and cluster scaling. Plus, it possesses a nice administration panel.
Not meant for free distribution.
Supported systems: Mac OSx, Linux and Windows
Benchmark: Can be solicited to Objectivity
Every time it’s more frequen for companies to offer hybrid solutions that use a variety of database engines to give entrance to various NoSQL models, and even to relational engines.
For example CortexDB, Foundation DB and Orient DB offer a variety of NoSQL models.
IBM has extended its DB2 database in order to offer the possibility to use NoSQL databases with BLU Acceleration. IBM DB2 in its extended version allows storing data in XML, JSON and graph mode formats.
Databases as a Service
Databases that offer their services on the cloud. You only need to read and write your data using services offered by the purveyor. Possibly these types of databases will be one of the most demanded because of its ease of use and low learning curve, along with the tendency to transfer to cloud services from small and medium companies.
This database, as it name implies, is meant for simple databases. If you want to make accessing your databases something easy, without complex queries such as comparatives, then Amazon SimpleDB might be a good choice for you.
Data is stored as text and structures formed by value-parameter peers are generated.
No benchmark available
List of Pandora FMS modules for database monitoring
Pandora FMS is a flexible monitoring software, capable of monitoring devices, infrastructures, applications, services and business processes. And, of course, it also has numerous modules and plugins for database monitoring. Let’s see some of them:
We hope this article has been useful for you. Remember, if you wish to add any database don’t hesitate to drop a comment.
Before saying goodbye, do you want to know better what Pandora FMS can offer you? Find out by going here .
If you have to monitor more than 100 devices, you can also enjoy a FREE 30-day Pandora FMS Enterprise DEMO . Get it here .
Finally, remember that if you have a reduced number of devices to monitor, you can use the Pandora FMS OpenSource version. Find more information here .
Do not hesitate to send us your questions. Pandora FMS team will be happy to help you!