Shulou(Shulou.com)05/31 Report--

This article will explain in detail how to analyze the impact of Oracle 18c on DBA and how to deal with it. The content of the article is of high quality, so the editor shares it for you as a reference. I hope you will have a certain understanding of the relevant knowledge after reading this article.

Obviously not, believe it or not, I want to tell you for sure that after going to the cloud, DBA will play a more important role in the enterprise.

I have been engaged in the DBA profession for 17 + years, and I have more profound experience and understanding in this aspect. Many friends are worried about whether they will lose their jobs in the future. first of all, let's take a look at the development trends of this industry:

1. There will be more fine and advanced features, as is the case with every new version.

2. In the database, more and more tasks can be completed automatically by the system, so it is very good for both enterprises and individuals to upgrade to the new version as soon as possible, rather than waiting for forced upgrade.

3. In the future, the requirements for DBA will be higher on the cloud.

So in this article, I will talk about the impact of the introduction of the Oracle autonomous database on DBA, and discuss with you how DBA should respond to new trends.

Oracle Database 18c is the next generation of industry-leading databases.

Oracle introduced the world's first autonomous database to OOW this year, and its corresponding cloud platforms and services achieve higher performance, security and reliability requirements at the lowest cost, reduce the complexity of operation, reduce the probability of human error, and most of the work can be done independently, reducing the workload of manual operation.

I would like to emphasize "Database Cloud" and "Oracle autonomous database cloud" here, because when we talk about databases on the cloud, Oracle's autonomous database cloud is actually a cloud database service. In this article, we will call it "cloud database"

The topic of autonomous database and cloud database can actually be analyzed from different angles. Most of the articles I have seen are about the advantages and benefits of this future database. So what aspects should we focus on?

1. Who will decide which service mode the database will be in?

2. Who will embed policy constraints into these qualification databases?

3. Who has enough knowledge about the routine tasks and behaviors of the database to decide how to reduce the cost of these services?

4. When we have more choices, the infrastructure of IT will become more and more complex. Who will decide the design of these systems?

Obviously, the answer to all these questions is DBA, however, not any ordinary DBA can do it. In order to accomplish these tasks, DBA must have an in-depth and comprehensive understanding of this future cloud database.

As I just said, autonomous database is actually a different kind of cloud database service. Therefore, the first thing to understand is how to migrate the database from the local to the cloud. For cloud migration of databases, please refer to: Oracle Cloud (DBaaS): Migrating Databases to Oracle Cloud Using RMAN Backup

The Oracle autonomous database is a database with many general tasks that have been automated, which can be automated as follows:

1. Application of patches

2. Upgrade

3. Independent optimization of the system.

However, the focus of this article has not yet begun:

First of all, I introduced the exact words of one of my ACED friends, Tim Hall, who said that the pre-sale of 18c has almost no impact on DBA, and only when the service suite of autonomous databases is launched as a whole will it have a greater impact on DBA.

How do you understand this sentence?

First of all, for the first half of the sentence, Oracle18c has no effect on DBA, it's just a higher version. It is not a general relational database management software running in autonomous mode, in fact, the autonomous database itself is designed for future environment and requirements, it is only for the cloud, it has nothing to do with the local database.

In the second half of the sentence, the service suite of autonomous databases has an impact on DBA. Autonomous database is a suite of services available on the Oracle public cloud. This means that local databases cannot be run in autonomous mode, which may be implemented later.

At present, many DBA are worried about whether the autonomous database service suite will put them out of work, but this is still a long way off.

In fact, there are several services that provide databases on Cloud:

1. Oracle database cloud service

2. Oracle bare metal cloud database service

3. Oracle database all-in-one cloud service

4. Oracle database integrated machine, cloud machine

5. Oracle database fast cloud service

The autonomous database service suite will represent other possible services that you can sign up for.

Next we will discuss what else we should know about Oracle's autonomous database:

1. Oracle autonomous database, or save my drive database, will be fully launched in the 18c version, which is very different from the current 12c version.

2. The version of Oracle12.1 should have at least 4 years left and is not expected to be eliminated before 2021.

3. Oracle12.2 should provide extension services in 2025. We all know that when a new version is launched, many users will not rush to upgrade the database to the latest version, but will not upgrade until the service that needs to respond or the new version of the extension service is about to expire. If you think about it this way, it will take a long time for Oracle18c to be put into production environment on a large scale.

At present, Oracle autonomous database is designed for Exadata, we know that Oracle Exadata is very powerful, but very expensive, so many users will not choose, especially for some small and medium-sized enterprises.

Therefore, DBA need not worry, at present, Oracle18c will not be completely autonomous, and autonomous databases will not completely replace the traditional data-based operating mechanism.

Next we are going to discuss several more important topics:

1. Oracle 18c is not an autonomous database service, and vice versa. These are two concepts

2. Autonomous database service components are currently only applicable to Oracle public cloud services.

3. According to the current situation, the autonomous database service component only supports the environment of Exadata. (of course it may change in the future.)

4. Oracle 18c is just a new version of the database

When we understand this, we can say with certainty that the introduction of autonomous databases does not have much impact on the local DBA of the current operation and maintenance. But that doesn't mean we don't need to change in the face of cloud trends and database trends. Only when we have an in-depth understanding of new technologies and directions, understand their advantages and disadvantages, and make preparations in advance, will we not be caught off guard by the new wave.

Next, let's talk about some of the most attractive features and features in Oracle's autonomous database. After all, the Oracle autonomous data warehouse cloud was launched in December this year.

Automatically apply patches: in the current situation, the process is simple if you want to apply a patch set to the database. Check the latest patch set on the official website and follow the installation documentation and instructions, which will be completed soon.

Therefore, it is expected that this kind of procedural manual operation will soon be realized by the automated program of the system.

There are also some patch sets that need to be stopped when they are applied, because the program will modify the binaries in the system. But in this case, Oracle probably already has a corresponding mechanism for automation. In fact, as long as the opinion task can be broken down into some sequential steps, it is possible to achieve it through system automation. Therefore, for the process work such as patching, it will naturally become one of the first tasks to be automated.

Upgrade: when using databse cloud service, the only way to upgrade a database running in the cloud is to create a new service, in which there is a dedicated compute node that we can use to upgrade the database. However, we should make it clear that in terms of PDB management, Oracle strives to establish a very advanced mechanism. For example, we can hot clone PDB and migrate PDB from one container to another without affecting business and operation. These functions are essentially similar to the principle of migrating data files online, but the level of implementation is more advanced, so we see that the technology of Oracle is becoming more and more mature.

Tasks such as upgrading can also be quickly defined as what tasks need to be done on PDB and what configuration needs to be done on CDB to ensure that the database will run properly after the upgrade. And I'm sure the work will be done online without shutting down the database. From this point of view, the automatic upgrade technology is essentially no different from the technology we are using on the local database, except that under a new service model, these technologies can be applied at a higher level.

In the following content, my ACED friend Tim helped me explain some elements so that you can better understand the Oracle 18c database. I quote him below: "Oracle has explained how the process of automatic upgrade and patching is implemented in 18c database, for 18c database running in Exadata environment, because 18c supports all processes of rolling upgrade and patching, including OJVM, online patching for services provided by Oracle."

Self-optimization: this sounds complicated, but it's actually a very simple principle. In the current environment, when we use some adaptive features in the database, the database is equivalent to self-optimization, such as automatically creating indexes, all of which are done online. At the same time, it is not very difficult to add the AI engine to the database to better collect and analyze the data, and then reflect it in the project of SQL query.

In other words, self-optimization is the use of similar adaptive features to influence the choice of SQL execution path after analysis by AI programs.

For some of the things that the above system can accomplish on its own, let's take a look at the frequency of execution:

1. Apply the patch set: apply the patch set is not a frequent task, hit it regularly, and the execution frequency is very low.

2. Upgrade: the frequency is lower, and the general database version is only updated every few years, but for most customers, they are not in a hurry to upgrade as soon as the new version is released, so there is less need for such an operation.

3. Self-optimization: the frequency will be very high, almost continuously, because the data changes in the database are very frequent, and the corresponding optimization will be used to add, delete, query and modify the data, that is to say, it will be very expensive to enable this feature. We know that in the current database, there is tuning advisors, and in our experience, the effect is not very good. In many cases, after we take the optimization advice given by advisor to adjust, the performance is even worse, so what will be the effect of automatic optimization in autonomous databases? If it is really perfect and can be well optimized in real application scenarios, it will indeed reduce the corresponding demand for DBA.

Therefore, a very important thing is, without 100% certainty, how much do you think an enterprise is likely to completely adopt systematic self-optimization without any artificial testing and control.

I think the possibility of this is very low, so there are many factors to be considered for optimization. Apart from SQL itself, we also have to consider the logic of the application, the design of the architecture, and even some policy restrictions, and so on. Very often, people cannot perfectly take into account every aspect of optimization, let alone machines.

Let's take a simple example. In some environments, Oracle Dataguard has an automatic failover mechanism. Sometimes some human errors in the database cause the database to switch failover automatically. In fact, we do not want to switch these scenarios. Therefore, in order to avoid the impact of automatic failover, many enterprises are afraid to use FSF (Fast Start Failover). Although this feature is very functional, it will always apply a lot of data that the system does not allow to be implanted in the system.

Generally speaking, I think that autonomous databases will greatly reduce the demand for DBA work, but it will not completely replace the existence and role of DBA.

Autonomous databases promise users the following advantages:

1. Reduce management time

In infrastructure construction, in upgrades and patches, in ensuring high availability, and in performance scratches, time will be greatly reduced

2. Increase the time for innovation

More time will be spent on data analysis, data policy, data security, and database design.

Therefore, after going to the cloud, DBA must enhance its security management skills.

So, the point is, what is the future of DBA in the face of Oracle's cloud database?

Seventeen years ago, when I first started working on DBA, it was easy to design a database architecture, just to decide what environment to install the database in, such as a server, a mainframe, or in certain scenarios, on a desktop machine, that is, PC.

Nowadays, there are many deployment environments for databases, such as servers, virtual machines, integrated systems such as Exadata, and many other options.

You must also decide what architecture the database will be embedded in, such as the most common local options, private cloud, hybrid cloud, integrated cloud, and with the introduction of 18c, options are growing.

So in this case, who decides to deploy the database in the above environment, in which service mode, or, of course, DBA. Therefore, it is not that DBA is not needed, but DBA is required to know more knowledge outside the system, business, platform and so on.

I divide the current Oracle DBA into the following three categories, and their directions are as follows:

The first category: daily work only focuses on some of the most basic and routine tasks, such as patching, capacity expansion, and so on. So when the autonomous database is launched, if they don't work hard for change, they are likely to lose their jobs.

The second category: while operating and maintaining the database, they also do other work related to IT, or have rich experience in other fields, then this kind of DBA can make more important decisions for the company through the comprehensive learning of all kinds of knowledge, not limited to the database. This is what we often call the transition from DBA to architect.

The third category: for those DBA who decide to go further in the field of Oracle, as the system becomes more and more intelligent and powerful, and the requirements for DBA become higher and higher, this kind of DBA needs to work hard to learn more knowledge to understand the business, the cloud, and all the skills needed in the cloud, so that it can make reasonable planning and design among the many choices of Oracle without getting lost.

On how to analyze the impact of Oracle 18c on DBA and measures to share here, I hope that the above content can be of some help to you, can learn more knowledge. If you think the article is good, you can share it for more people to see.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.