• +86-135-5029-3426
  • sales@vinchin.com

Vinchin Blog

Vinchin Blog A case of upgrading the core system database server of a group financial company

A case of upgrading the core system database server of a group financial company


< pre > Inspur commercial machine pre sales technical support, Bai Hongtao
1. Project background
The client of a group financial company (hereinafter referred to as the financial company) is approved by the China Banking Regulatory Commission and is subordinate to a group company. It is a company with the purpose of strengthening the group's capital management and improving the efficiency of capital use, giving full play to the overall advantages of the group company, optimizing the allocation of resources, reducing the cost of capital and improving the efficiency of capital, A non bank financial institution that provides professional and high-quality financial services for the internal members of an enterprise group. Its main business scope includes:
L handle financial and financing consultancy, credit assurance and related consulting and agency business for member units;
L assist member units to realize the receipt and payment of transaction funds;
L approved insurance agency business;
L providing guarantee for member units;
L Handle entrusted loans and investment between member units;
L handle bill acceptance and discount for member companies;
L handle the internal transfer settlement between member units and the design of corresponding settlement and liquidation scheme;
L absorbing the deposits of member units;
L to handle loans and financial leasing for member units;
Engage in interbank lending.
L underwriting corporate bonds of member units;
L equity investment in financial institutions;
L portfolio investment
In recent years, with the continuous development and growth of various undertakings of a group company, the optimization and adjustment of its capital layout structure is accelerated, the pace of market-oriented mergers and acquisitions is accelerated, the number of member units is increasing, and the characteristics of business mix and diversification are more and more obvious. The use of scientific and technological innovation and information construction to help the development of the group has become the key strategy of the group. In order to cooperate with the group's development strategy and better serve the group and its member units, the construction of the information system of the financial company is also increasing investment. To create a safe, stable, reliable and efficient core infrastructure platform has become the focus of the information construction of the financial company.
2. System status and demand analysis
The core business system of the financial company is the financial management service platform system. Through the integrated interface platform, the system realizes the external links with various systems, such as the people's Bank of China, commercial bank, banking regulatory bureau, enterprise credit reference, etc., and the internal links with the financial system, so as to provide financial services to all members of the group. The front-end of the financial management service platform system uses x86 server with virtualization software to run the application systems, and the back-end uses two IBM p720 minicomputers purchased in 2013. It is equipped with 8core power7 processor and 64GB memory to run Oracle database, which includes account management, fund settlement, loan business, bill management, fund budget, financing management, etc With the acceleration of group merger and acquisition, business function segmentation and operation years increasing, the financial management service platform system with IBM p720 minicomputer as database server is facing severe challenges, which are mainly reflected in the following aspects:
The financial management service platform system is the most important core system of the financial company, and the database server is the key hardware in the system. The IBM p720 minicomputer has been running for many years, the hardware is aging, and the stability and reliability are declining year by year.
IBM p720 minicomputer has been out of production for many years, and far beyond the official maintenance period. At present, it mainly relies on third-party maintenance, so it is difficult to guarantee the quality of spare parts and the timeliness of hardware problem solving.
At present, IBM p720 minicomputer adopts gigabit network interface and 8GB HbAS interface, and its bandwidth and data throughput capacity are low, which can not meet the requirements of business development.
With the accelerated development of the group's business, the business pressure of the financial management service platform system has increased significantly in the past few years. The performance of the existing host has reached a bottleneck. The utilization rate of processor and memory is very high at the end of the month during the peak period of settlement business, with occasional stuck and delay.
In the current system, only IBM p720 + powerha is used to ensure the high availability of local services, and there is no effective disaster recovery design. When there are room level failures such as power system failure, fire and other accidents, it may cause serious consequences such as business interruption and data loss.
In order to meet the challenges and better serve the group and its member units, the customers finally set up a project to upgrade and replace the database server of the financial management service platform system after several rounds of investigation and verification.
3. Overview of upgrade scheme
In this upgrade and replacement, the customer uses two latest generation key business servers K1 power s922 produced by Inspur Commercial Machinery Co., Ltd. to replace the existing IBM p720 server as the database server of the financial management service platform system, deploys the AIX 7.2 operating system and Oracle 11g, and uses powerha as the high availability cluster for the database to ensure the local high availability of the database, Two Inspur as2600g2 hybrid flash storage and dual live cluster are used to replace the existing v3500 storage, and IBM p720 server and v3500 storage are migrated to the disaster recovery room. Oracle ADG is used to do disaster recovery for the production environment database to ensure the business continuity and data security in case of room level failure.
After the upgrade, the topology of the customer's hardware platform is as follows:
4. Scheme highlights
O low risk in implementation process: K1 power s922 and IBM p72
0 is the same power architecture, which realizes the perfect undertaking and smooth migration of the existing business. In the migration process, it ensures the minimum impact on the production business, the lowest risk and the shortest downtime window
O powerful performance: K1 power s922 provides 4 times the overall performance of the current IBM p720, and 4 times the memory capacity of the current p720. It can easily cope with the current business pressure, break the performance bottleneck, and greatly improve the computing power of the whole machine
O strong scalability: K1 power s922 can provide stronger scalability, and there is still up to 3.75b memory expansion space left at present, which can easily cope with future business growth
O strong integration capability: born into the cloud, built-in enterprise virtualization software can improve resource utilization through virtualization technology, allow LPAR + PowerVM to realize physical partition + virtualization deployment, and provide strong integration capability
O significant improvement of business continuity: K1 power s922 can provide the strongest hardware and system RAS features in the industry. With powerha high availability cluster, it can completely solve the problem of old system hardware and declining business continuity. In addition, as2600g2 dual active storage architecture can ensure that the production business will not be affected in case of single storage level failure
O make full use of the old: the original p720 and v3500 equipment are used and migrated to the disaster recovery environment to ensure that the database can be quickly restored to the serviceable state in case of computer room level failure, give full play to the resources of p720 and v3500, and protect the investment
O cost reduction: 3-year 7x24 hours of original factory maintenance, spare parts quality and timeliness of fault resolution without worry, reduce maintenance costs,; Smaller volume than p720, saving production room space cost
O long life cycle: power minicomputer is stable and reliable, which has been verified. It can stably support the operation of the system for more than 7 years with one purchase, and its online service life is very advantageous.
5. Customer revenue
In this project, the customer replaced and upgraded the existing IBM p720 by two K1 power s922, which completely solved many pain points of the current financial management service platform system, seamlessly connected the software upgrade of the core system, and made sufficient preparations for the future development and reform of the group in the information system infrastructure cloud
1) Front end application experience improved significantly
Before the update and replacement, the processor utilization rate of IBM p720 was more than 90% at the end of each business peak, and the front-end business operation would be jammed and delayed occasionally, which seriously affected the work efficiency. After the update and replacement, the processor utilization rate of K1 power s922 remained below 20%, and the front-end operation had no jamming and delay, and the business departments and member units had good feedback.
2) Business continuity and data security improved significantly
Before the update and replacement, the number of database server and storage hardware failures gradually increased, and the business interruption time also extended. The backup of business production data is mainly done by the database administrator on a regular basis, and the backup files are copied to the offline storage media. RTO and RPO are difficult to guarantee. After the update and replacement, after several rounds of disaster recovery drills, the backup of business production data is completed, In the case of equipment single point failure and machine room level failure, RTO and RPO meet the requirements of the group, and the confidence of the operation and maintenance team is greatly increased.
3) Fit the future development direction of infrastructure
Although the customers did not use virtualization and cloud technology in this upgrade, the built-in enterprise virtualization software of K1 power s922 can help customers quickly transform to the construction direction of IAAs private cloud platform in the future, without replacing hardware equipment or purchasing virtualization software products repeatedly.
4) TCO revenue of core system database server
According to the customer calculation, through this upgrade and replacement, three years of third-party operation and maintenance costs and spare parts purchase costs are saved on the core system database server, the economic losses caused by unplanned downtime are reduced, and the space occupation and energy consumption of the production room are reduced. Considering comprehensively, the total cost of ownership of upgrading and replacement is better than continuing to use IBM p720.

  • Tag:
  • Cloud

Interested Blogs More