October 6, 2016 at 15:38Easy to copy - quickly restore
Solutions in the field of backup and data recovery are necessary for both large companies and representatives of medium and small businesses. However, the choice of how to use such solutions depends on the size of the organization and the challenges it faces.
Hewlett Packard Enterprise's flagship backup solution is HPE Data Protector. This product has a solid, almost 30-year history. Developed by Apollo in 1987 under the name Omniback Network Backup System, after Apollo joined the Hewlett-Packard in 1989, it developed for a long time and in 2002 received a new name - HP Data Protector. Despite the venerable age of the product, HP, and now Hewlett Packard Enterprise, is constantly working to improve it, reflecting in it all the technological changes taking place in the software world and in the IT industry as a whole. We will briefly talk about the main features of Data Protector, focusing on the most important and useful innovations implemented by developers in recent years.
Infrastructure backup unit (cell) consists of several components - the managing server and client parts. The control server part can be placed on any operating system: Microsoft Windows, Linux, and HPUX - Unix in the version of HPE. In the control part there is a database that contains information about the entire backup process: from what equipment it was carried out, what and when was reserved, etc. HPE Data Protector control servers can be clustered to ensure disaster tolerance, in addition, centralized support their management, which is especially useful and relevant for organizations with a geographically distributed structure.
The client part is represented by Disk Agent and Application Agent integration agents that transfer data to the Media Agent, which backs them up to standard disk arrays, network devices, physical and virtual tape libraries, as well as specialized storage systems with deduplication support. Data Protector supports both HPE StoreOnce proprietary storage systems and many third-party hardware solutions. And it’s not about simple data placement, but about integration, which allows, for example, to carry out distributed deduplication, centralized management of backup units and much more.
As the resources to be backed up, cloud environments can be used, not only public clouds, but also HPE Helion - a specialized solution based on the OpenStack open platform that allows the customer to build their own private cloud.
Management of HPE Data Protector is carried out through a simple and convenient graphical interface, as well as through the command line, and depending on current tasks and user preferences, you can switch from one method to another.
We should also dwell on some important functions implemented in the Data Protector in recent years. So, in the 8th version of the product, released in 2013, a completely updated internal database appeared, which is based on the PostgreSQL DBMS. Thanks to the use of a powerful industrial relational DBMS, the HPE Data Protector significantly accelerated, and in addition, the marginal possibilities of a solution in terms of the number of supported devices, simultaneous sessions, etc. increased many times. Thus, even large companies with extensive corporate IT infrastructure now one backup cell may well be enough, and several distributed cells are needed by those distributed organizations whose departments or divisions are either not connected at all by a common scarlet or enjoying an unstable connection.
One of the key features of HPE Data Protector is integration with HPE Store Once storage where backups are stored. This integration enables data deduplication across various sections of the IT infrastructure. Firstly, at the place of their creation, that is, on the client side. Of course, this will lead to an increase in the load on the client equipment, therefore, an option is provided when deduplication is carried out first on a dedicated server, and only then deduplicated data is transferred to storage. Finally, deduplication can also be done right on the HPE Store Once. All three options can be used within one backup infrastructure, if necessary, switching from one to another using the Data Protector.
As for backup applications, HPE Data Protector integrates with most software solutions and databases from Microsoft, VMware, SAP, Oracle, IBM Lotus, Sybase, etc. Integration is carried out using application agents. Since all the fine-tuning of the backup of these systems is carried out directly in the Data Protector, the work of the system administrators and IT specialists responsible for the backup is greatly simplified - you do not need to spend time setting up the backup options in each application.
To protect virtual environments, HPE Data Protector integrates with the most popular hypervisors - VMware and Microsoft Hyper-V, although the basic copying capabilities of other virtualization solutions are present. At the same time, both copying and recovery can be carried out on the basis of both LAN and SAN networks. In addition, a few months ago, a separate product for backup virtual environments, HPE VM Explorer, appeared in the Hewlett-Packard Enterprise product portfolio. This affordable and simple solution may well be in demand by small and medium-sized enterprises.
However, copying is only half the battle, because the data needs to be quickly restored. By integrating HPE Data Protector with many devices and environments, including cloud, you can provide a combination of different levels of recovery. An analogy with the placement of passengers and their belongings on an airplane is appropriate here. There is a business class, and somewhere else the first one, as the highest, then the economy class and, finally, the luggage compartment. In our case, the more important the data, the faster it can be restored, but the recovery process will have an appropriate price.
The fastest and most expensive level is an instant recovery from hardware snapshots (system snapshots) on disk arrays. Such a hardware snapshot is created on the disk simultaneously with the backup and stored there. The capabilities of HPE Data Protector allow you to organize the rotation of snapshots on a disk array and get several points for recovery at once. This will be useful when you need to “raise” a critical database from a backup in a matter of seconds: you don’t have to restore anything in the usual way, just switch volumes on a disk array.
Virtual machines also do not need to be restored in the classical way, since they can be started directly from the backup, and then transferred to the main disk in the background.
For a number of systems, the so-called granular restoration is provided, which allows you to restore certain elements, and not the entire system. In particular, separate directories and files can be restored in VMware virtual machines, and separate objects can be restored in Microsoft Exchange and SharePoint: a letter, a file, etc.
Many customers will want to use the bare metal recovery function, that is, to a completely clean server , without preinstalled OS and other software. To do this, you must prepare the bootable ISO-image of the server in advance, placing critical software on it that may be required for quick recovery in case of serious hardware malfunctions.
Starting with version 7, HPE Data Protector supports backup monitoring through integration with IT infrastructure monitoring tools such as HPE Operations Manager and Microsoft System Center Operations Manager. Then there was a new HPE Backup Navigator module responsible for backup analytics. For example, it will tell you when a backup device is full, the quota on the disk array is over, etc. This tool can generate over 70 different types of reports, and IT specialists can present them to management as a justification for modernization and expansion. hardware backup infrastructure. It’s important to note that HPE Backup Navigator works with its own database,
A useful addition is the new built-in task scheduler, which sorts backup sessions according to the established priorities. Thus, more important sessions will be the first to gain access to the freed device. In addition, the scheduler displays a list of failed backup sessions and automatically starts them.
In conclusion, a few words about licensing policy. HPE Data Protector has two licensing schemes - by component and by volume of protected data. The component licensing scheme allows you to select a list of licenses individually for a particular infrastructure, taking into account the number of necessary integrations with application servers, used tape drives, disk devices, and other options. The licensing scheme for the volume of protected data makes it possible to simplify the selection of HPE Data Protector licenses by calculating them according to the total volume of protected data. For various infrastructures, one or another licensing scheme may be more profitable, therefore, it is recommended to request a calculation for both schemes - this is easy to do, because HPE has a Russian-language questionnaire,
Hewlett Packard Enterprise's flagship backup solution is HPE Data Protector. This product has a solid, almost 30-year history. Developed by Apollo in 1987 under the name Omniback Network Backup System, after Apollo joined the Hewlett-Packard in 1989, it developed for a long time and in 2002 received a new name - HP Data Protector. Despite the venerable age of the product, HP, and now Hewlett Packard Enterprise, is constantly working to improve it, reflecting in it all the technological changes taking place in the software world and in the IT industry as a whole. We will briefly talk about the main features of Data Protector, focusing on the most important and useful innovations implemented by developers in recent years.
Infrastructure backup unit (cell) consists of several components - the managing server and client parts. The control server part can be placed on any operating system: Microsoft Windows, Linux, and HPUX - Unix in the version of HPE. In the control part there is a database that contains information about the entire backup process: from what equipment it was carried out, what and when was reserved, etc. HPE Data Protector control servers can be clustered to ensure disaster tolerance, in addition, centralized support their management, which is especially useful and relevant for organizations with a geographically distributed structure.
The client part is represented by Disk Agent and Application Agent integration agents that transfer data to the Media Agent, which backs them up to standard disk arrays, network devices, physical and virtual tape libraries, as well as specialized storage systems with deduplication support. Data Protector supports both HPE StoreOnce proprietary storage systems and many third-party hardware solutions. And it’s not about simple data placement, but about integration, which allows, for example, to carry out distributed deduplication, centralized management of backup units and much more.
As the resources to be backed up, cloud environments can be used, not only public clouds, but also HPE Helion - a specialized solution based on the OpenStack open platform that allows the customer to build their own private cloud.
Management of HPE Data Protector is carried out through a simple and convenient graphical interface, as well as through the command line, and depending on current tasks and user preferences, you can switch from one method to another.
We should also dwell on some important functions implemented in the Data Protector in recent years. So, in the 8th version of the product, released in 2013, a completely updated internal database appeared, which is based on the PostgreSQL DBMS. Thanks to the use of a powerful industrial relational DBMS, the HPE Data Protector significantly accelerated, and in addition, the marginal possibilities of a solution in terms of the number of supported devices, simultaneous sessions, etc. increased many times. Thus, even large companies with extensive corporate IT infrastructure now one backup cell may well be enough, and several distributed cells are needed by those distributed organizations whose departments or divisions are either not connected at all by a common scarlet or enjoying an unstable connection.
One of the key features of HPE Data Protector is integration with HPE Store Once storage where backups are stored. This integration enables data deduplication across various sections of the IT infrastructure. Firstly, at the place of their creation, that is, on the client side. Of course, this will lead to an increase in the load on the client equipment, therefore, an option is provided when deduplication is carried out first on a dedicated server, and only then deduplicated data is transferred to storage. Finally, deduplication can also be done right on the HPE Store Once. All three options can be used within one backup infrastructure, if necessary, switching from one to another using the Data Protector.
As for backup applications, HPE Data Protector integrates with most software solutions and databases from Microsoft, VMware, SAP, Oracle, IBM Lotus, Sybase, etc. Integration is carried out using application agents. Since all the fine-tuning of the backup of these systems is carried out directly in the Data Protector, the work of the system administrators and IT specialists responsible for the backup is greatly simplified - you do not need to spend time setting up the backup options in each application.
To protect virtual environments, HPE Data Protector integrates with the most popular hypervisors - VMware and Microsoft Hyper-V, although the basic copying capabilities of other virtualization solutions are present. At the same time, both copying and recovery can be carried out on the basis of both LAN and SAN networks. In addition, a few months ago, a separate product for backup virtual environments, HPE VM Explorer, appeared in the Hewlett-Packard Enterprise product portfolio. This affordable and simple solution may well be in demand by small and medium-sized enterprises.
However, copying is only half the battle, because the data needs to be quickly restored. By integrating HPE Data Protector with many devices and environments, including cloud, you can provide a combination of different levels of recovery. An analogy with the placement of passengers and their belongings on an airplane is appropriate here. There is a business class, and somewhere else the first one, as the highest, then the economy class and, finally, the luggage compartment. In our case, the more important the data, the faster it can be restored, but the recovery process will have an appropriate price.
The fastest and most expensive level is an instant recovery from hardware snapshots (system snapshots) on disk arrays. Such a hardware snapshot is created on the disk simultaneously with the backup and stored there. The capabilities of HPE Data Protector allow you to organize the rotation of snapshots on a disk array and get several points for recovery at once. This will be useful when you need to “raise” a critical database from a backup in a matter of seconds: you don’t have to restore anything in the usual way, just switch volumes on a disk array.
Virtual machines also do not need to be restored in the classical way, since they can be started directly from the backup, and then transferred to the main disk in the background.
For a number of systems, the so-called granular restoration is provided, which allows you to restore certain elements, and not the entire system. In particular, separate directories and files can be restored in VMware virtual machines, and separate objects can be restored in Microsoft Exchange and SharePoint: a letter, a file, etc.
Many customers will want to use the bare metal recovery function, that is, to a completely clean server , without preinstalled OS and other software. To do this, you must prepare the bootable ISO-image of the server in advance, placing critical software on it that may be required for quick recovery in case of serious hardware malfunctions.
Starting with version 7, HPE Data Protector supports backup monitoring through integration with IT infrastructure monitoring tools such as HPE Operations Manager and Microsoft System Center Operations Manager. Then there was a new HPE Backup Navigator module responsible for backup analytics. For example, it will tell you when a backup device is full, the quota on the disk array is over, etc. This tool can generate over 70 different types of reports, and IT specialists can present them to management as a justification for modernization and expansion. hardware backup infrastructure. It’s important to note that HPE Backup Navigator works with its own database,
A useful addition is the new built-in task scheduler, which sorts backup sessions according to the established priorities. Thus, more important sessions will be the first to gain access to the freed device. In addition, the scheduler displays a list of failed backup sessions and automatically starts them.
In conclusion, a few words about licensing policy. HPE Data Protector has two licensing schemes - by component and by volume of protected data. The component licensing scheme allows you to select a list of licenses individually for a particular infrastructure, taking into account the number of necessary integrations with application servers, used tape drives, disk devices, and other options. The licensing scheme for the volume of protected data makes it possible to simplify the selection of HPE Data Protector licenses by calculating them according to the total volume of protected data. For various infrastructures, one or another licensing scheme may be more profitable, therefore, it is recommended to request a calculation for both schemes - this is easy to do, because HPE has a Russian-language questionnaire,