CHALLENGES FOR DATA RECOVERY

Few will doubt that, just as the great technological advances of the last five years have increased productivity, they have also increased the difficulty associated with data recovery. A problem that, together with the new and robust enterprise storage solutions, is exacerbated exponentially.

Today all organizations manage huge amounts of structured and unstructured data in traditional databases, virtual environments and in the cloud. With storage in multiple locations and high availability, information is constantly moving between users, making it difficult for companies to know where their information is at any given time. Added to this is the possibility of human errors that accidentally delete information. Together these factors make recovery very difficult in the event of a data loss.

Today, the main challenges for data recovery in the most common storage systems are the following:

Taming the cloud

Many organizations have gone too fast to relinquish control of their most critical information to cloud providers. Unfortunately, many still do not know what problems they can suffer and how is the recovery process in case of data loss.

There are two common situations that clearly aggravate data recovery in cases of loss. The first is given because despite the fact that a company's information is isolated, many clients access the same physical storage space. When a specific customer data loss or failure occurs, service providers sometimes cut off access to the environment, trying to protect other customers. The second occurs because many vendors use their proprietary virtual or storage machines, and do not share details about the system configuration to protect their intellectual property, which prevents research and development of the correct solution.

It is essential that companies understand the importance of entering into a service level agreement (SLA) with their supplier and know its limits before the loss occurs. The best guarantee is that the cloud storage solution provider works with a trusted data recovery company, offers comprehensive services, and is able to minimize downtime caused by storage failures or even errors on the part such as accidental erasure of virtual disks. In this way, customers will know where and how their information is stored, and how it will be retrieved if a loss occurs.

The headache of virtualization

Computer failure, item deletion or file corruption are some of the common data loss cases. But there is a real challenge when it comes to data recovery in virtual environments: Being made up of a physical machine joined by multiple virtual machines, if the physical device fails it can cause other machines to fail, increasing the impact of data loss. In addition, finding the parts of the information and putting it back together is difficult, because the information is fragmented on the storage platform and in constant motion. If we add to this that there may be small files, we are faced with a real recovery puzzle.

Virtual environments contain many groups of data, and the key to not overloading your storage is balancing load capacity. The user has the opportunity to place large amounts of information in a single storage environment, but the challenge arises when it comes to recovering data at this scale, and having the tools to fully recover it and return it to the client correctly. In addition to the amount of information, less fragmentation leads to a higher success rate. Large amounts of data make it difficult to find the individual fragments or specific virtual disks required to remount damaged or deleted virtual machines.

Solid state hard drives

Solid state hard drives, SSDs, based on flash, offer great benefits such as high speed (low read latency, random access, and start-up time), low consumption, lightness, silent, and very resistant to shocks. This, added to the fact that the cost per gigabyte is falling rapidly, has skyrocketed their adoption and the amount and value of the data they store, making future data loss catastrophic for the businesses or customers involved.

Many believe that SSDs are immune to data loss due to lack of moving parts, compared to traditional equipment (HDD). However, although they are less likely to be damaged by a drop, they can suffer data loss and their unique characteristics make recovery very complex.

In the most extreme case, SSD data recovery can be time consuming due to the need to investigate which algorithms were originally used to store it. With SSDs, the location of the data changes every time it is rewritten, making recovery much more difficult. Also, depending on the complexity of the device and its capabilities - error correction code (ECC), garbage collection, split volume techniques, compression, encryption, etc. - the task of reordering the information is complicated, making the process can take from a few days to several weeks. SSDs are still in the early stages of their technology life cycle and vary greatly from one manufacturer to another and between different ranges of the same manufacturer. Sometimes they can even be different within the same model! The variations are mainly due to changes, improvements, and firmware updates that manufacturers carry out to improve the operability of devices and to meet consumer demands; however, this process adds difficulty and complexity to recovery.

It is important that, in addition to knowing the challenges presented by this technology, standards and recovery solutions are developed in parallel with the evolution of the SSD with the aim of instilling confidence in storage “integrators” and customers, to adopt and improve the technology of solid state hard drives.

 

What is in your database?

Database recovery, be it from damage to physical equipment, internal database corruption, or data erasure, is not as straightforward as you might think. Each database is unique and complex, with an internal structure different from others, with different versions and improvements that appear constantly. Data recovery providers must follow all of these formats and enhancements to make recovery a success.

Additionally, corrupted, deleted, or missing data can create a number of recovery problems, requiring a thorough analysis of the complex internal structure of the database. When a storage device is down, or a file system needs to be repaired, many companies think it's impossible. Although it is not, it requires that the database files be rebuilt from raw fragments. With all these possible scenarios, the recovery method should allow consumers access to their data when it has been recovered.

Another challenge that arises is that of physical data versus logical data. When hardware fails, many data recovery providers attempt to recover only information at the physical level, but ignore the logical level. Providers need to be able to pull sensitive consumer data from the virtual tier, as well as physical data from various systems and file configurations, for full, quality recovery.

 

Conclusion

When an organization has its data, whether structured or not, in many different locations, recovery in the event of loss becomes more complicated. Whether hosted in a traditional database, solid state hard drive, virtual environment, or in the cloud, each presents a number of challenges when attempting recovery. Therefore, it is recommended that both companies and individuals speak openly with their storage providers to know where their data is at all times. In addition, they must ensure that in the event of data loss, they know the protocol of the recovery process and the procedures to follow to minimize the impact on the business environment and save the confidential data of the company.