The fact that every reality must have one or more data protection, backup and disaster recovery solutions is a concept that has now been seen and revised and that in this article we take for granted, especially in light of the events of the last period that have affected various companies in the global IT sector.
The objective is to compare the main differences between two backup methods on the market and focus mainly on Agentless backup through a concrete and innovative example such as Storware.
Backup Agent or Agentless?
There are mainly two possible approaches for performing backup and recovery operations: with Agent or Agentless. Both solutions have their potential and limitations, so it's essential to understand the difference and choose the best solution according to the needs of each organization.
Backup with Agent
Agent-based backup involves installing an agent to perform all tasks; the latter becomes responsible for managing the backup process and communicating with the backup server itself.
This method certainly provides a level of scrutiny on granular data and the possibility of configuring more complex backup operations; it guarantees greater visibility in the backup process, allowing monitors and manage the backup more effectively; finally, it improves the performance of the backup itself, reducing the network load.
This does not mean that it is also good to consider its limitations such as dependence on the operating system and surrounding activities such as updates, management and maintenance of the infrastructure itself. Agent backups also require resource consumption and related set-up costs on each machine. Sometimes they can also complicate the management of the backup infrastructure.
Agentless Backup
As the name suggests, it is a backup method that eliminates the need to install an agent on the systems to be protected, making the backup strategy and process easier to implement, significantly reducing the time and effort for monitoring it.
In fact, by choosing Agentless Backup you get simplicity in the installation and configuration of agents on individual systems; suppleness relating to different operating systems and platforms; scalability in managing large amounts of resources and systems, reducing the impact on performance itself.
On the other hand, compared to agent backup, Agentless backup has possible limitations due to the lack of advanced features (usually provided by the agent) and a longer wait for data to be requested through the network, this could in fact take longer.
The choice between these two types of Backup essentially depends on the needs of each reality, however, as we told you, in this article we want to learn more about the Agentless method.
Our opinion on Agentless Backup
As in other situations, at CloudFire we love to get involved and learn about new technologies or approaches, and it is for this reason, the time has also come to get out of balance by giving you our opinion regarding the Agentless backup topic.
In our opinion, the advantages of this technology are different, here are the main ones:
- Consistency of applications: Agentless backup, which provides a complete point-in-time version of a server or VM, allows you to restore the workload, ensuring the continuous execution of applications without loss of data;
- Easier administration and management: Agentless systems allow administrators to oversee each workload, even in the most specific aspects of the entire network. In other words, when the backup managers set up the network to be used and the data required in the backup, the latter are migrated directly from the network into the defined repository;
- Effective control: Policy management is a common feature of Agentless backup, thus promoting better control of backup and recovery processes, while reducing working time;
- Reduction of setup and maintenance costs: Agentless almost always translates into lower costs, specifically it cuts the costs related to the purchase of an agent license for each workload on the network, and also reduces CPU and bandwidth to a minimum.
Storware: Agentless backup for Openstack
Our infrastructure and ours Public Cloud are based on Openstack: one of the most complex opensource projects and at the same time suitable for creating, managing and distributing a pool of compute, storage and networking resources in a scalable and modular way.
Already in itself, the structure of an Openstack environment requires high skills and experience, and it is therefore preferable that a solution be used for protection and backup that does not increase difficulty or effort.
At CloudFire, we choose Storware as a Backup & Recovery solution because:
- It's Agentless.
Its distinctive feature is precisely related to the fact that it does not require the installation of any agent, facilitating the complexity of data protection in an environment like Openstack for the reasons that we have already explained at the beginning of the article. In fact, Storware is able to speak directly with the infrastructure through the OpenStack APIs; - It supports different Storage Repositories.
Storware supports multiple repositories for volumes and storage, including Object Storage such as Ceph and Storage compatible with S3, just like our Object Storage, to which Immutability can be added. In fact, Storware supports incremental backups and the transfer of disk images directly from the API; - Reduces effort for system engineers
Those who use the instances no longer have to worry about installing and managing the agent on the guests; it is the system administrators who take care of the backups without having to have any operating system credentials; - It makes available more ways to back up VMs in a virtualization environment.
From Disk Image Transfer for KVM hypervisors that use the QQOW2 image format, creating a copy of the entire disk image of the VM; to connecting the disk through Cinder, connecting the VM disk to a Proxy server dedicated to backup, and copy a copy of the VM disk.
In conclusion, the protection and protection of each infrastructure component in our case in Openstack environments is fundamental.
Storware is the first technology that can offer this service in Agentless mode, guaranteeing extreme versatility both for the supported platforms and for the target repositories, and all this fully combines the values and characteristics shared by CloudFire.
