Home>Article> Goodbye MySQL, it may no longer be a sensational meme!

Goodbye MySQL, it may no longer be a sensational meme!

青灯夜游
青灯夜游 forward
2022-11-09 10:04:03 2021browse

In recent years, everyone may have this feeling: Unlike the programming language market, the competition in the database market is extremely fierce - the first-tier companies are stagnating or even declining, while the second-tier companies are overtaking. There are various signs that MySQL, the biggest brother on the popular list, is gradually fading out of the sight of professional developers. Goodbye MySQL, it may no longer be just a sensational meme!

In August, the 2022 developer survey report released by StackOverflow included this set of numbers:

In the popularity rankings among professional developers, PostgreSQL (46.48%) surpassed MySQL for the first time (45.68%) came out on top. Professional developers differ from beginners in that they are more likely to choose Redis, PostgreSQL, Microsoft SQL Server, and Elasticsearch.

Among the beginners, MySQL is not optimistic, followed by MongoDB, which ranks second, accounting for nearly 1/3. "This makes sense because it supports a large number of languages and application development platforms."

It is worth mentioning that developers who are using MySQL are also considering, hoping or wanting to work with other databases. The chart below shows the survey results.

Goodbye MySQL, it may no longer be a sensational meme!
Source: StackOverflow

There are 11,185 MySQL developers who want to work with PostgreSQL, and 9,520 MySQL developers who want to work with MongoDB.

Not only the StackOverflow report, but also based on the latest October DB-Engines popularity trends, it can be seen that the popularity of MySQL has been declining for several consecutive years.

Goodbye MySQL, it may no longer be a sensational meme!
Picture source: DB-Engines

It can be judged that if nothing unexpected happens, it is only a matter of time before PostgreSQL, which is also an open source relational database, surpasses MySQL.

1.MySQL is stagnating

As the "king", MySQL's crown of glory was crowned in the era of the Internet explosion. Over the past 20 years, scale and efficiency have become the unspoken "essentials" of the Internet's expansion.

After development and iteration, the database is simply used as storage and the computing tasks are handed over to the business server. This kind of logical deployment is more convenient to implement. Companies and engineers have recognized "Linux Apache MySQL PHP", and "LAMP" has naturally become the benchmark model for Internet development.

MySQL, which had just been born at that time, was born in a simple style for the "minimalist" Internet at the right time. Although from a functional point of view, MySQL is not as dominant as other databases, but because it does not provide many advanced functions, MySQL, which is simple to use, open source and free, has won the favor of domestic Internet companies including Tencent and Alibaba that later became giants. Goodwill became the de facto standard in the early Internet OLTP field. In comparison, PostgreSQL, which has always followed the "academic" route, has always been tepid.

2009 was the year when MySQL ushered in a turning point in its destiny. Oracle made a move on SUN, which had just bought MySQL. In the face of the rise of MySQL, Oracle has to say that this move is intended to adopt an acquisition strategy to protect the moat of Oracle, a large commercial database. Subsequently, Oracle began a price increase routine: significantly raising the price of the commercial version of MySQL.

Like developers who use the free version of MySQL, MySQL founder Monty is worried about its future. He is filled with indignation, but there is nothing he can do.

Even after more than ten years, Monty is still concerned about this matter, and MariaDB is the best proof.

Looking back, the rise and popularity of MySQL can never be separated from the context of the times. As Monty said:

The success of MySQL is inseparable from the background of the times. At that time, the Internet had been widely recognized, and everyone needed such a database to use it to create the data needed for the Internet. At that time, technology giants took a wait-and-see attitude towards the Internet, so this was a blue ocean market that had yet to be developed.

2. Behind the stagnation

When the blue ocean turns red, MySQL, which was born for the Internet, seems to have ushered in a moment of stagnation. Although the popularity has not diminished, compared with the popularity of PostgreSQL, a contemporary relational database competitor, which has bucked the trend, it is inevitably not worrying.

Upon closer inspection, there are probably several reasons:

First of all, after the acquisition, the open source spiritual core of MySQL was greatly reduced. Being reduced to the acquired "post-son" will inevitably become a marginal role.

According to Monty's recollection, "When Oracle announced that it would acquire Sun and MySQL, I did not believe that they would truly explore the valuable legacy of MySQL, so many outstanding programmers decided to leave together with us and continue The spirit of MySQL."

At the same time, the outstanding employees who joined Oracle because of MySQL also did not receive enough resources and attention.

In December last year, Steinar Gunderson, the former chief software engineer of Oracle and a member of the MySQL optimizer team, resigned. He criticized MySQL, which he had devoted five years of hard work to, and said: "There is only so much you can do. I The changes he and the rest of the team made moved the MySQL optimizer toward its early 21st century design with some nice tweaks, but that was the end of it." No matter how much bragging was done within the company, Gunderson couldn't believe MySQL. will become a competitive product.

Perhaps for the members of the Oracle MySQL optimization team, the lack of room for imagination is an indescribable pain. Of course, Oracle has also spent energy on MySQL, but patchwork "operations" such as Microsoft Oracle Cloud and online analysis and processing capabilities are far from the ideal of a true open source database contributor.

In short, Oracle's acquisition of MySQL may not stop its research and development, but it has greatly sapped the enthusiasm of the development community.

Secondly, the context of database development has changed, and MySQL’s highlight window period may really be coming to an end. As mentioned before, MySQL was born for the new network at that time, and the development of major websites and mobile terminals had also reached an insurmountable point. Today, we see that industry development is beginning to tilt towards traditional industries such as finance, telecommunications, the Internet of Things, retail, and manufacturing. Compared with the Internet twenty years ago, these industries pay more attention to data reliability, security, and standardization. These are not what MySQL is good at.

Another point is that since the epidemic, the modernization requirements of enterprise infrastructure have been accelerated and put on the agenda, aiming to make enterprises more flexible and responsive to rapidly changing customer needs. The global system integrators who undertake these projects often choose to apply the easiest to deploy technology to bring the best profits to their services. This is obviously not MySQL. The reason is very simple. MySQL requires payment, and MySQL Multi-language support obviously has its own bottlenecks.

Finally, competing products are everywhere. As far as relational databases are concerned, the market is also undergoing significant changes. Since 2014, PostgreSQL has ushered in a renaissance. Technical analysts attribute the reasons to four points: rich feature set, easy expansion, open source, and better open source license. The last point seems to the author to be crucial. The implications behind better open source licensing are huge. The greater the scale of open source collaboration, the more everyone benefits. This is also the core of the open source spirit.

PostgreSQL adopts an MIT-like license agreement, allowing developers to do anything, including commercial use in open source or closed source products, while the MySQL client follows the GPL license agreement, so developers must pay Oracle or transfer Open source your own applications. From this perspective, PostgreSQL is the most advantageous choice whether it is for commercial use or otherwise.

3. PostgreSQL that takes the academic route

MySQL will not become more popular until the next breakthrough comes. However, before that, the second rise of PostgreSQL was only a matter of time.

Less than 14 months after MySQL was released, PostgreSQL was born. It is an advanced open source RDBMS developed by the PostgreSQL Global Development Group. It was originally released on July 8, 1996 and started in 1986 at the University of California, Berkeley, as part of the POSTGRES project.

When PostgreSQL was born, it was regarded by developers as an "academic" with "rigorous design concepts". To use a popular saying nowadays: MySQL is a wild Internet solution to solve a certain business problem, while PostgreSQL is a "full-stack database" that is "one against ten".

Born to be a hyper-converged system, mature, well-designed, professional and inclusive community, PostgreSQL confidently advertises itself as "the world's most advanced open source relational database", data analysis, timing plug-in , full-text search, architecture and functions, etc. are all good at it.

Earlier we analyzed the reasons for MySQL’s stagnation from the perspectives of open source enthusiasm, industry development, and commercial use. Here we list some of the benefits of PostgreSQL from a developer's perspective:

  • Supports a variety of performance optimizations that can be used for commercial solutions, including geospatial data support, read lock-free concurrency, etc. , is widely used in large systems;

  • is most beneficial for systems that need to perform complex queries;

  • performs well in business intelligence applications , more suitable for data analysis and data warehouse applications that require fast read/write speeds, therefore, it is also suitable for OLTP/OLAP systems;

  • Can store structured and Unstructured data type, which supports most data types, such as support for JSON. One of the biggest innovations in PostgreSQL over the years is the introduction of the ability to generate JSON data in its PostgreSQL 9.2.

Powerful functionality may not be the reason for popularity. But it’s true that developers’ preferences have changed.

For example, James Governor, an analyst at Redmonk, an industry analysis company focusing on software developers, mentioned that “developers currently have some NoSQL and big data fatigue.” Therefore, developers began to use the time-tested PostgreSQL as MongoDB and A viable alternative to Apache Cassandra for some critical workloads.

In addition, Elijah Zupancic, director of solution engineering at cloud computing company Joyent, also mentioned the importance of documentation, “PostgreSQL has also been recognized by developers. From their perspective, using it is a pleasure. , the documentation is excellent and the data types reflect the type of work developers do.”

PostgreSQL is not dependent on any single vendor. A single commercial vendor will never be able to keep up with the speed of change that open source projects can offer at levels like documentation updates.

4. Simple and easy to overwhelm everything

Looking back at the development of MySQL, it was born because of the Internet and also made the Internet possible. In that era, "simple" and "easy" were overwhelming. Technology is changing rapidly, but the logic behind this is still applicable today.

PostgreSQL may not be the best choice, but for business decision-makers, it is the most convenient and easiest one. Many within the enterprise are accustomed to relational databases, and PostgreSQL is an "easy button" for managers who want to abandon expensive commercial databases.

As EDB CEO Ed Boyajian mentioned, most companies don't want to upgrade and change, but they are using PostgreSQL as their greenfield because they already have the SQL/relational skills in-house, and those skills It's accumulated over decades of using Oracle, SQL Server, and DB2.

5. Goodbye, MySQL!

In the final analysis, MySQL has completed its mission of building the Internet because it is easy to use; and with the proposal of new enterprise development propositions: digital transformation of traditional industries, this Including several sub-propositions such as "upgrading infrastructure", "moving to the cloud", and "moving to the Internet of Things", MySQL's shortcomings are highlighted. This is not an area that MySQL can easily handle. Whether you are an integrator or a developer, it is not difficult to understand why you should choose PostgreSQL, which is more mature, rigorous, well-designed, and easier to succeed.

Goodbye MySQL, it may no longer be a sensational meme!
Picture source: StackOverflow

In the past nearly 20 years, MySQL has become one of the cornerstones of building the Internet. Countless developers have worked day and night for it, and we have given it Due glory. Nowadays, developers have to face the problem of MySQL: it has long passed the peak of experience, but is facing a sense of stagnation or even decline.

What is the future of databases? More stable and more dynamic. Monty said so.

Looking forward, enterprises and developers are faced with new stable and dynamic needs, and they have to embrace the future. They have to shout out the phrase that was once a joke but is now a prophecy at some point. "Sigh of taste - Goodbye, MySQL!

Statement:
This article is reproduced at:51CTO技术栈. If there is any infringement, please contact admin@php.cn delete