Cloud Server Backup
 |
| Cloud Server |
Preparation of a new cloud server for work to begin with a backup. All seemed to know about it - but sometimes even experienced system administrators admit unforgivable mistakes. And it is not only that the task of setting up a new server needs to be solved very quickly, but even that is not always clear what backup method you want to use.
Of course, the ideal way that would suit everyone, cannot be created: all have their pros and cons. But at the same time it is quite realistic to choose the method most appropriate to the specifics of a particular project.
Read More About: Free VPS
When you select a backup should first pay attention to the following criteria:
1. Speed (time) backup in the vault;
2. Speed (time) restore from backup;
3. How many copies can be kept with a limited amount of storage (storage server backups);
4. The amount of risk because of the inconsistency of backups, not regulated method perform backups, complete or partial loss of backups;
5. Overhead: the level of the load created on the server during backup, reducing the response speed of service, etc.
6. Rental price of all services used.
In this article we will talk about the basic methods of cloud server backup and the most common problems that may be faced newcomers in this very important area of system administration.
Scheme of the cloud server storage and retrieval of backups
When selecting schemes of backup method should pay attention to the following basic points:
1. Backups cannot be stored in one place with redundant data. If you store a backup copy on a disk array to your data, you will lose it in the event of damage to the primary disk array.
2. Mirroring (RAID1) cannot be compared with the backup. Reid will only protect you from a hardware problem with one of the drives (and sooner or later this problem will be, because the disk subsystem is almost always the bottleneck on the server). In addition, when using hardware raid there is a risk of breakage of the controller, i.e., you must keep it a spare model.
3. If you store your backups in a single rack in DC, or just under one DC, then in such a situation, too, there are certain risks (you can read about, for example, here.
4. If you store your backups in a different DC, the sharply rising costs of the network and the speed of recovery from a remote copy.
Often the cause of data recovery is the damage of the file system or disk, i.e. backups should be stored somewhere on a separate server storage. In this case, the problem may be the "width" of the data channel. If you have a dedicated server, the backup is highly desirable to perform on a separate network interface, not the same, that performs data exchange with customers. Otherwise, your customer's needs may not "fit" in the limited communication channel. Or because customer traffic backups will be made in time.
Next you need to think about the scheme and the recovery time from the standpoint of storage backups on Cloud Server. Maybe you can be quite happy with that backup is performed within 6 hours at night in the vault with limited access speed, but the restoration of length at 6 o'clock you are unlikely to want. So access to the backups should be comfortable and the data to be copied quickly. For example, the restoration of 1TB of data with a bandwidth of 1 GB / s will take almost 3 hours, and it is if you do not "rested" in the performance of the disk subsystem in the storage and server. And do not forget to add to this time a problem is detected, the time for a decision on a pullback, while checking the integrity of the recovered data and the amount of the subsequent discontent clients / colleagues.
You Might Be Interested in: Cloud Computing