Existing UNIX servers of various manufacturers generally have high reliability. In this regard, IBM's P series servers are particularly prominent. But all UNIX servers cannot achieve the reliability level of IBM mainframe S / 390, which is determined by the architecture and application environment of the open platform server. Using IBM high-availability cluster software-HACMP, can better protect critical business applications from failures.
How HACMP works
HACMP is an abbreviation of High Availability Cluster Multi-Processing. HACMP is IBM's highly reliable cluster software on the P series AIX operating system. It is configured with redundancy, eliminates single points of failure, and guarantees continuous availability and safety and reliability of the entire system. HACMP is to use the network to detect the status of the host and the network card. With the hard disk mirroring and other functions provided by AIX, when the host, network card, hard disk control card, hard disk or network fails, it automatically switches to another set of spare components to work ; If the host fails, switch to the backup machine to continue the operation of the application system.
Two servers as a dual-machine system run HACMP software at the same time; there are three backup methods for the two servers: one server runs the application, and the other server is used as a backup for the two servers. The other party's backup host; two servers running the same application at the same time, the need for database support. The two host systems monitor each other's operation status (including the system's software and hardware operation, network communication and application operation status, etc.) through the "heartbeat line" during the entire operation process; Once the other host's operation is found to be abnormal (failure) , The application on the faulty machine will stop running immediately, the machine (the backup machine of the faulty machine) will immediately start the application on the faulty machine on its own machine, and the faulty machine application and its resources (including the used IP address And disk space, etc.) to take over, so that the applications on the failed machine continue to run locally; the process of taking over applications and resources is automatically completed by HACMP software without manual intervention; when the two hosts are working normally, one of them can also be The application on one machine is manually switched to another machine (backup machine) to run.p5 520 HACMP full redundancy recommended solution:
Schematic:
Two IBM p5 520 servers are used to run the application, and HACMP software is run separately to ensure the high reliability of the system.
Two 2005-H08 fiber switches were used to establish a storage LAN environment, and two p5 520 servers and disk arrays were connected respectively. 2 fiber switches can avoid single point of failure. Using IBM DS4300 to ensure the reliability of data storage and reading efficiency. Two p5 520s are connected to two storage fiber switches through two fibers, and the DS4300 is connected to two storage fiber switches through four fibers. This connection ensures reliability and improves data access efficiency.Recommended configuration:
2 1.5GHz / 1.65GHz power5 processors, 4GB memory, 2 73GB internal hard drives: used to install the operating system 1 Ethernet card Typically, a server should provide at least two network cards (a service network card and a backup network card) . Configure an Ethernet card. Another network card uses the built-in Ethernet card. One fiber card is connected to the fiber switch. One 2-port asynchronous card provides a serial port to connect the heartbeat cable. Provides a point-to-point connection between two nodes, used to transfer control information and heartbeat information when a TCP / IP subnet failure event occurs HACMP software AIX operating systemFiber optic switch IBM Total Storage SAN Switch 2005-H08:
8-port fiber switch for connecting servers and disk arrays.
Disk array IBM Total Storage DS4300: used to store user data
YDS Product categories of Silicone Gifts , we are specialized manufacturers from China,Silicone Bracelet & Rings
1.Mold workshop and 2D and 3D engineer department
2.Solid siliccone compression machine and liquid silicone injection machine
3.Disney and Sedex 4P audit factory
4.ISO 9001,IATF16949,Raw material of FDA LFGB MSDS Certificates
Look forward to your cooperation!
Silicone Gifts,Silicone Sport Jewelry Wristband,Silicone Carering Cover,Silicone Cartoon Luggage Tags
Shenzhen Yindingsheng Technology Co., Ltd , https://www.ydsoemsilicone.com