Shulou(Shulou.com)06/01 Report--

[Enmo College] it turns out that banks are using these databases.

Old fish

Inscription

Different from the widely used open source MySQL database in the Internet industry, banks have higher requirements for availability and security, which must be taken as the premise of any innovation and business. So what kind of databases are banks using? Today, this article will reveal the secrets for you one by one for selection reference.

In this issue, Lao Yu covers as many different types of banks as possible in the interview for your reference. The interview banks include not only the four major banks, the Bank of China and the Industrial and Commercial Bank of China, but also the China Merchants Bank, Pudong Development Bank and Jiangsu Bank of the joint-stock banks.

Big data has become a national strategy, and the first problem big data faces is big data's storage problem, which can not bypass the database, because the database is used to store data application software. Any database has its advantages and disadvantages, whether it is traditional relational database or NoSQL database or NewSQL database. Therefore, the decision-making process of which database to choose is very complicated for enterprises.

Different from the open source MySQL database widely used in the Internet industry, banks have higher requirements for usability and security, which must be taken as the premise of any innovation and business. at the same time, mobile banking and online banking also have the characteristics of large customer volume, large transaction volume and high transaction peak value (such as great promotion). In most cases, banking business should meet the requirements of ACID, and there can be no data illusion. All these bring great challenges to database selection, architecture, performance, operation and maintenance. So what kind of databases are banks using? Today, this article will reveal the secrets for you one by one for selection reference.

(special note: some of the respondents belong to departmental limitations, therefore, the enterprise business lines listed below are not comprehensive, only major application databases are listed for reference.)

Industrial and Commercial Bank of China

Industrial and Commercial Bank of China (ICBC) is one of the four major banks, with the largest customer base in China, with 5.784 million corporate customers and 530 million individual customers at the end of 2016. For the fourth year in a row, it topped the list of the world's top 1000 Bankers, the Forbes Global 2000 and the Fortune 2000 Commercial Banks.

Interview object: Wu Hui, the head of the ICBC data center platform database, is responsible for the daily maintenance of more than 10 kinds of databases and thousands of sets of databases on the open platform. responsibilities mainly include all kinds of database architecture deployment, system launch, high availability construction, fault diagnosis, upgrade and withdrawal and other lifecycle management operations.

Based on the commercial database, ICBC has independently built a full range of bank application systems, covering personal, public, credit card and other business lines, hundreds of application systems.

There are many types of databases, including traditional commercial databases, including DB2, Oracle, SQL Server, etc., and open source databases such as MySQL, etc.; there are both relational databases, unstructured platforms such as Hadoop and Spark, and distributed cache platforms based on Redis to supplement relational databases.

The core business systems of ICBC are mostly above DB2 and Oracle. In terms of open source MySQL applications, ICBC focuses on promoting the widespread use of artificial intelligence, the Internet of things and other innovative areas, and matches the characteristics of the bank to carry out a number of innovations in architecture deployment, parameter tuning and other aspects, which has become the key direction of the follow-up transformation of OLTP relational databases.

Industrial and Commercial Bank of China started the construction of enterprise data warehouse a long time ago, and constructed an enterprise data warehouse system based on Teradata products to process and summarize hundreds of applications and provide data to the downstream.

Based on Hadoop, spark and other open source large database products, ICBC independently constructed a big data platform to deal with unstructured data. At present, there are more than 700 nodes, and the amount of data reaches 6PB, providing data support for more than 40 applications, such as online banking, financial e-couplet.

Bank of China

The Bank of China was officially established on February 5, 1912. It is the only bank in China that has been operating for more than a century, and it is also one of the four major banks in China. In 2016, it ranked sixth in Banker's Top 500 Global Bank Brands list and 35th in the World's Top 500 Brands list.

Interview object: Zhang Haibin, senior manager of database operation and maintenance team of Bank of China data center. The work includes: daily database inspection, problem solving, routine guarantee and key periods, important project guarantee, external service vendor management, data security, architecture design and adjustment, performance optimization, key and difficult problem solving, etc. it also includes the introduction of new technology, innovation, personnel training and so on.

Bank of China has a more diversified choice of databases, including commercial databases Oracle, DB2, SQL Server, open source databases MySQL, MongoDB, Hbase, and domestic database tide K-DB.

Among them, DB2 is mainly used in the mainframe (MainFrame) and used in the core system of the bank.

Open platform (UNIX) platform is mainly Oracle database, involving a large number of peripheral systems, including key channel system, bus system, report processing system and accounting system, and so on.

The application on the Win platform is mainly related to the front end of the branch, mainly using SQL Server. In the application of x86 platform, there are many kinds of databases, such as Oracle, MySQL, MongoDB,Hbase and so on.

Bank of China also has a lot of innovation in application, which mainly embodies the following aspects: in the aspect of automation, there are automatic installation and configuration of database, automatic health check, and building database problem troubleshooting platform. it also includes performance data warehouse and data mining. Build the main database and standby database environment, optimize the data structure and application architecture. Combined with ADG technology to achieve high data availability guarantee, part of the application system to achieve read-write separation, optimize the implementation of application functions, and improve the efficiency of database services. Combined with the data replication technology of storage equipment, it provides the Tunable 1 data query function of part of the system.

China Merchants Bank

China Merchants Bank is the first joint-stock commercial bank in China which is wholly owned by an enterprise legal person. It was founded by Hong Kong Merchants Group Co., Ltd. on April 8, 1987. It is the sixth largest bank in mainland China and one of the eight banks and five guarantees for Chinese financial stocks in Hong Kong.

Interview object: Tian Yongjiang, Manager of Database and Application Management Room of China Merchants Bank data Center. Responsible for database and application planning and construction and daily operation and maintenance management.

China Merchants Bank database types are also more, and the most commonly used are Oracle, DB2, MySQL, respectively used in important business systems or high-complexity business systems, as well as ordinary systems.

Tian Yongjiang said: "the innovation and development of financial technology is very fast, and it is our constant pursuit to protect the business, and even to promote and lead the business development, and it is necessary for us to better understand and grasp the development trend of the industry." do a good job in technology introduction and digestion. "

Perhaps for secret-related reasons, he did not talk about the specific innovation of China Merchants Bank, but he also pointed out that no matter which kind of database, as long as it has the requirements of availability, performance and capacity, it will carry out architectural planning and design, such as sub-database, read-write separation, and multi-activity, and deploy in two places, combining disaster recovery and one-click switching to form a highly available, multi-center deployment, and scalable architecture.

Shanghai Pudong Development Bank

Shanghai Pudong Development Bank (Pudong Development Bank for short) is a state-controlled national joint-stock commercial bank established on August 28, 1992, opened on January 9, 1993 and listed on the Shanghai Stock Exchange in 1999. Its head office is located in Shanghai. Shanghai Pudong Development Bank ranked 49th among the top 2016 Chinese enterprises in August 2016.

Interview object: Yang Xinjie, database architect of Information Technology Department of Pudong Development Bank, responsible for database architecture management and design.

Pudong Development Bank has a more diversified choice of databases, including commercial databases such as Oracle, DB2, Informix, SQLserver, Sybase, data warehouse and OLAP: Teradata and GreenPlum. There is also the open source database MySQL,Redis,Hbase.

Among them, Oracle is mainly used in credit card business in Pudong Development Bank. Pudong Development Bank has fully embraced the 12c multi-tenant system and is promoting DBaaS in an all-round way. In the future, all databases are deployed on a few clusters (database resource pools). The goal is to achieve dynamic scalability, rapid deployment and self-service. DB2 is embracing the Purescale architecture in its entirety.

While Informix and Sybase are gradually withdrawing. MySQL is being gradually piloted and promoted on some peripheral and channel systems. Hbase is also used on big data platform.

Jiangsu Bank

Jiangsu Bank is a modern joint-stock commercial bank formed by merger and reorganization on the basis of 10 urban commercial banks in Wuxi, Suzhou, Nantong and other cities in Jiangsu Province. It was officially opened on January 24, 2007. It is the only provincial local corporate bank in Jiangsu Province. By the end of 2016, total assets reached 1.5983 trillion yuan, deposits totaled 907.4 billion yuan, and loans totaled 649.4 billion yuan.

Interview object: Zhu Tao, the head of the system management team of the Information Technology Department of Jiangsu Bank, is responsible for the management and maintenance of hardware and database, middleware and other software in the production and disaster recovery center of the head office.

There are also many types of databases in use in Jiangsu banks, mainly Oracle, as well as emerging open source and distributed databases. At present, MySQL, Gbase, MongoDB, Redis, Hbase, GoldenDB and TDH are all used on a small scale in Jiangsu banks.

Among them, the Oracle database is mainly used in the traditional OLTP environment application in Jiangsu Bank. For the traditional Oracle database, Jiangsu Bank has changed the traditional mode of applying a set of database, and has carried on a large number of integration of the database. In the last three years, the new Oracle database is all running on the PC server RAC cluster, and each RAC carries multiple sets of applications. And continue to migrate the old stand-alone version of the database to the cluster environment, the number of stand-alone databases continues to decrease.

Different from other banks, in the database used by Jiangsu Bank, we see more domestic databases, in which Gbase database is a domestic column distributed database commonly used in Nanjing University, which is mainly used in part of the analytical system of Jiangsu Bank, and GoldenDB is the distributed relational database of ZTE, which is currently used in part of the background system of Jiangsu Bank. The TDH of Shanghai Star Ring has been widely used in data marts, customer tags, flow computing and other aspects in Jiangsu Bank.

Summary

In fact, during the three-month interview, Lao Yu interviewed far more people from banks and industries. Generally speaking, banks are generally diversified in the choice of databases. Most of the important business systems or high-complexity business systems are based on traditional commercial databases such as Oracle and DB2. Ordinary systems will also choose open source databases such as MySQL and domestic database products.

The survey found that IBM DB2, which used to occupy an absolutely dominant position in the banking industry, was not so easy, and the open platform DB2 was fine, but the mainframe DB2 was greatly affected by de-IOE, coupled with the market strategy that IBM did not do much for DB2. Although DB2 products were excellent, they still could not resist the decline of market share, and there was a downward trend from the first echelon to the second echelon. Oracle CEO Mark Hurd mentioned not long ago, "if it were so easy to migrate databases, DB2's market share would probably be zero." Although this was made in response to Amazon, it made a mockery of the growing marginalization of DB2.

Therefore, if IBM does not have more new measures for DB2, it may be sooner or later to be eliminated, but the market position of DB2 will not change much in a short time. In addition, there is a factor that can not be ignored, the introduction of DB2 by banks does not rule out the consideration of business checks and balances.

Sybase was hidden by SAP after it was acquired, and there was little sound in the market. according to the survey, the situation was bad, and the only bank in use in the interview also had plans to eliminate it.

It is worth noting that we have found that there are many domestic databases in the banking industry, such as Nanjing University General Motors, Tide, Star Ring and ZTE. Although its products are not used in the core system of the bank, at least this is a good start, which means that the domestic database has made great progress in usability and security, and achieved a breakthrough in the banking industry. As for whether the domestic database can go further in the future, we will wait and see.

Apart from banks and state-owned enterprises, what is the trend of the database in the whole market? Traditional SQL cannot cope with massive data and high concurrency scenarios, but NoSQL cannot be used because of lack of standards. at present, there are many NewSQL distributed databases on the market. A hundred schools of thought contend, where will the database go in the future? please refer to the database competition: who is the king of distributed NewSQL from SQL to NewSQL?

Reproduced from: Lao Yu's notes

Original: http://mp.weixin.qq.com/s/7qujkM1kDEst7SZRhhhuAg

Enmo College, a subsidiary of Yunhe Enmo (Beijing) Information Technology Co., Ltd., is committed to providing professional and high-level oracle database and big data training services, mining and training big data and database talents. Enmo College provides a full range of big data and database technology training, including individual practical skills training, personal certification training, and enterprise internal training. ACE-level super teachers, equipped with professional laboratories, immersion learning and training, professional laboratories, equipped with professional teaching assistants to guide training. Can quickly integrate into the circle of experts, rich in resources in the industry, and quickly accumulate workplace contacts. Oracle database courses include: Oracle DBA practical class, Oracle OCM examination, Oracle OCP examination and so on.

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.