Automated backups with customized archiving
Backups are the last line of defense to prevent permanent data damage and loss. Backups also decrease downtime in the event of a disaster. Despite the critical data protection provided by backups, a significant number of production servers still lack this vital defense.
ServePath provides a range of automated backup systems. Used in conjunction with our network storage product, managed backup allows customizations best suited for your server's environment and business needs. In the event of a disaster, you will be able to rapidly restore your server and data.
Benefits of managed backup:
1. Backups are configured to run automatically, without any user
intervention.
2. Each backup is logged with a timestamp, so you can see how
long your backups take.
3. Archives of your data are made from backups and are available to you on
our network storage servers.
Technology Overview
ServePath offers two types of backups - full and differential.
Full Backups
With full backups, a new copy of the directories you provide will be made daily and stored onto our network storage servers. This type of backup offers the highest success in data recovery. Developmental servers and other dynamic environments will benefit most from full backups. You will have constant access to the last seven days' of backups, and all the Sunday backups from the last month, for a total of 11 backups (maximum 100 GB of archive storage is included).
Differential Backups
With differential backups, your server computationally assesses the changes on your file system since the last backup, and sends only the updates. This type of backup is best for production environments, where configuration and software updates are minimal, and the only consistent changes are to access logs, error logs, or databases. You will have constant access to the current week's sync of your server, as well as the Sunday sync of your server for the last month, for a total of 4 backups (maximum 100 GB of archive storage is included).
What are managed backups?
Managed backups are a product available from ServePath with a monthly charge for customers on our network. ServePath will automate the backup process to your customizations (see options below), and configure the server so that it is backed up nightly with a fixed archiving policy. In the event of a disaster, you can either contract ServePath for the time and materials to restore your data, or we can make the data available to you free of charge.
What are the different types of managed backups?
ServePath offers two backup implementations: differential and full.
A differential backup computationally assesses which files have been changed in a file system since the last backup, and only sends the delta, or changes, made since the time of the last update. Differential backups have traditionally been used in production environments, where code base and/or configuration updates are made incrementally in a stable environment. While there is a higher CPU overhead associated with computing the delta, the overall backup process is much lighter relative to that of a full backup. ServePath will create a full backup from the current week’s differentials, and archive the backups for a month, for a total of four backups plus the nightly updates.
| Differential Backup Diagram: | Full Backup Diagram: |
 |
 |
A full backup is a newly created copy of every file in the directories provided. This backup will be archived weekly for a month, for a total of four backups plus the last weeks’ backups. Full backups are most efficient for servers in a developmental or dynamic state. If applications and/or files get added and removed on a consistent, regular basis, then the full backup maximizes the probability of successfully restoring all your data in the event of a disaster.
What are some typical scenarios for managed backups?
Scenario 1 – Production server
You have a production server that typically does not see much in the way of changes to the file system. Access and error logs will grow and shrink as they are rotated out of use, and the production database will grow and shrink, but for the most part, everything on the server is stable, and there will not be any additional software installed or removed. Projections indicate that the environment will not undergo any drastic changes.
This is a classic example of an instance where differential backups are ideal. Given that this server is not undergoing any active development or drastic changes, a differential backup can be trusted to be accurate and will suffice for disaster recovery.
Scenario 2 – Pre-production server
You have a development server which has been promoted to pre-production status. The last development milestone has been reached, and the system is now in a usable state. However, due to the politics of deploying a production server, there may be last-minute changes that will ultimately mandate what additional software will be installed or removed.
This is a borderline case for differential backups. While we can ensure data integrity due to the end of development, drastic changes are still possible. If we continued with differential backups, and you decide to implement some last-minute changes that include either the installation of software and/or de-installation of software, then there is a high chance of data corruption.
Scenario 3 – Development server
You have a development server where a group of programmers are actively updating the system. Changes are made almost daily, with many changes rolled back due to incompatibility with other parts of the system.
This is a classic case for full backups. Differential backups should not be used here, as drastic changes to the file system (such as installing and removing software) tend to be problematic while computing the changes since the last update. Full backups will be more efficient for this environment, and maximizes the probability of success during disaster recovery.
Scenario 4 – post-development server
You have a server on which the development has stopped while testing and Q&A is happening. Barring any Q&A snafus, this server is ready to go into production status.
This is a borderline case for full backups. Any quality assurance process could very easily localize an issue to a specific piece of software and/or code, which will have to be changed or removed to ensure operations. Full backups, while incurring the highest overhead, are the best bet for any environment where significant change can occur.
How much data can I store?
ServePath's Managed Backup service includes 100 GB of storage. For full backups, 11 copies are retained (7 daily and 4 weekly), which means that the total amount of data that can be backed up is 9 GB. For differential backups, 5 copies are retained, which means that the total amount of data that can be backed up is 20 GB. Additional storage is available in 100 GB increments.
Please also note that full backups are written first to your local server and then copied across, so your server needs free space equivalent to the amount of data you wish to back up. In most cases, differential backups are recommended as they provide complete data protection with less space required.
What if I need to restore data from the backups?
ServePath's professional services team can restore data from any previous backup point to your server for $75 per half hour (one hour minimum) or $125 per half hour emergency.
What data is backed up?
ServePath's Managed Backup service is designed to protect your critical data. It does not make an image or snapshot of your drive, but in some cases it can be used to back up configuration files. For a high availability or rapid disaster recovery capability, however, we recommend multiple servers and some of our replication solutions.