Administrator Essentials Mainframe Backup Tool Application Guide

Among all the utilities available on the mainframe, those tools for backup are usually the most popular. What is the reason? You have to start with the basics of mainframe backup and regular backup.

After the mainframe crashes, there are two recovery methods: rollback/roll forward (it uses backup) and backup/restore. When using the rollback/roll forward method, you need to track the record transaction operation in the log. During the recovery process after the crash, the transaction is rolled back or revoked until the state reaches "stable" (does not contain invalid or inconsistent data, for example, Each backup has a stable state, and then roll forward the transaction until it is very close to the system state at the time of the crash. When using the backup/restore method, you need to take a snapshot of the system periodically, including all the data, and then save them on a dedicated backup storage medium (usually using tape or disk). When recovering, just take the latest system snapshot ( Backup version) Copy back to the system.

The purpose of backup is to minimize the loss caused by system crash. In addition, the backup operation should also minimize the occupation of system resources, because after the system crashes, in the process of restoring to the state before the crash, whether it is Whether you roll back a transaction or copy a snapshot, you must minimize the impact on performance.

Since the 1960s, the amount of data and the amount of data that needs to be backed up has grown at an alarming rate. It is estimated that the current global data storage has reached ZT bytes, and the development speed of backup processing capability is far behind. Not the growth rate of data. In addition, the system online requirements have also changed from the original working day (24*5) to 24*7. Many times, backups cannot be performed at all, because the amount of data is too large, and the IT budget can only meet the growing storage demand. Backup requirements can only be temporarily left open.

Therefore, it is the current top priority to develop a reasonable backup strategy. The backup strategy must be comprehensive and cover all the data as much as possible. It is necessary to do a backup job and reduce the backup cost as much as possible. Segmentation, such as structured data, semi-structured data, and unstructured data, then optimize backup strategies for each data type.

The mainframe provides comprehensive support for backup, it can handle hundreds of virtual machines, a large number of disk and tape storage, but in order to achieve fine-grained backup control, IT departments must automate the mainframe The backup tool performs DIY.

Mainframe backup tools are usually associated with batch jobs. At the beginning of the mainframe, users find that backup operations may be delayed. Wait until the end of the day to back up all the data in one batch. At this time, the backup operation will not be interrupted by other processes, so the backup task is extremely fast, which means that no other process will modify the data during the backup. Therefore, the weekend and weekend nights become the best time for backup. During this time period, the system can be offline (cut off the user's connection and avoid the impact of other processes on the backup operation). The backup operation is performed specially. At this time, the backup tool needs to be modified. 24*7 operation. At the same time, storage and database vendors also provide their own backup solutions, but they must be combined with the main online/offline backup tools of the mainframe. If you have seen the mainframe utility guide, you know that all backup tools actually Storage replication operations are performed offline, and each backup tool balances and coordinates other backup processes without causing backup conflicts.

Let's take a look at the mainframe's IEBCOMPR and IEBCOPY backup tools. IEBCOMPR will first compare whether the two files are the same, and then decide whether to back up, but the finer-grained backup needs to check in addition to checking whether the files are the same. Whether the file has changed, if there is no change, it will not be backed up. Therefore, IEBCOMPR can be used to determine whether the data on the online storage and the backup storage are consistent, and then decide whether to back up. Another quick way is to check the system log to see if there is any The transaction modifies the data, but most of the time, this information is not visible in the system log, so smart administrators will use IEBCOMPR as a scalpel instead of an axe.

IEBCOPY looks like a file copy command, but in fact it has more features. It can also merge data sets. On the surface it looks like it is copying data on online storage, actually in the copy process. It also squeezes out the redundant parts between the two data sets, which takes up less space on the backup storage. In other words, after running IEBCOMPR or checking the system log to determine which files do not need to be backed up, IEBCOPY can further eliminate redundant data at the file block level, so that the data that needs to be backed up is drastically reduced.

Although in the short term, the backup space can be solved to solve the problem of insufficient backup space, but in the long run, IT departments should focus on developing and implementing a fine-grained backup strategy to reduce backup data. Quantity is the right choice. In any case, mainframe backup tools such as IEBCOPY and IEBCOMPR have an important place in the administrator's command list. The flexible use of these backup tools not only improves backup efficiency, reduces backup space, but also greatly reduces backup requirements. time.