Importance Of Centralizing Your Data Storage
It is inevitable that when companies start looking into solutions to backup/restore their data, the concern of privacy and security immediately comes to mind. Data storage centralization was thus born as a cost-effective and affordable solution in overcoming this universal challenge. Since it’s emergence, the concept has evolved in such a manner that even an ordinary file server can change into a multifunctioning device with the right packages installed onto its operating system.
Many compliance audits usually recommend hardening as the process in securing the organization’s mission-critical data, but this option may not always be available or easily available for access if a cloud service was used for cold storage. It also opens a can of worms that may serve to be more trouble than it’s worth to tackle these new vulnerabilities. At which point data storage centralization becomes the ideal solution when considering the cost, manageability, and reliability.
Data also becomes easier to manage when kept in one place, and troubleshooting is less painless than opening a ticket and waiting for a service representative to investigate the issue. That means tighter control over data protection and enhanced security. Regardless of the circumstance, there will always be a single, consistent set of data. It also helps to reduce expenditure and risk while allowing greater flexibility over one’s hardware, performance, and storage capacity. Contrary to the idea that centralizing data in a single location is a security risk and an accident waiting to happen, scattering data all over the place increases the number of portals through which intruders can compromise. Ergo, protecting a central server is not only easier but also a more cost-effective approach of doing business as opposed to decentralization.
Access Control List
Synology’s ACL, (or Access Control List) allows an administrator to fine-tune the privacy of each user’s’ ability to view and open specified files and folders through the GUI console of the DSM OS. This can be extremely useful if you want to lockdown in which user is able to view what and the contents that reside within them.
To put ACLs in layman terms, they are essentially key cards that grant controlled access to certain places within a building. For example, John, an employee of “Yorkshire Loafers” is given an access card with a tag that reads “Human Resources”. He will only be able to unlock the door to the “Human Resources” Department. If he tries to use the card to access the “Sales” Department of “Yorkshire Loafers”, he will not be authorized for entry.
RAID and what they are
RAID is pivotal in the grand scheme of data fault tolerance. In the event of an unforeseen disaster, there must be a contingency plan in place to be able to recover the data that has been corrupted or lost. RAID (Redundant Array Of Independent Disks) is a category of data storage virtualization technology that takes advantage of multiple drives stored in the same chassis and combines them into a single large volume with different levels of redundancy and performance. Levels of performance, storage capacity and reliability is dependent on the volume type.
You might now be asking, how do I identify the best volume type to ensure the best levels of performance, stability, and redundancy for my setup? Luckily Synology offers a vast away of volume types to choose from. Let us briefly touch upon the different volume types that Synology offers in their NAS line.
Note: Certain RAID volume types are not available in the 2-Bay NAS of both the PLAY and + models as they require >2 drive slots to function properly.
Synology Hybrid RAID
Synology Hybrid RAID, or SHR is a proprietary offering that facilitates the storage deployment process. It was designed for beginners in mind to protect their data from hardware failure, without having to delve too deeply into complex subjects like parity and striping. Unlike the traditional RAID, SHR doesn’t have multiple volume types but rather they come in two different versions. SHR and SHR2. If 1 HDD gets damaged or corrupted, SHR provides the ability to protect data with the other HDD remaining intact. SHR-2 allows for the protection of data loss with 2 damaged or corrupted HDDs but this is only applicable to 4-Bay and above models.
A Basic volume type configuration is comprised of 1 HDD as an independent unit. It would be akin to having an external HDD hooked up to a router. This configuration does not provide data redundancy.
JBOD, or officially termed “Spanning” refers to a set of HDDs that hasn’t been configured to utilize the RAID system to improve data redundancy and performance. It combines a collection of HDDs that are stored within the array to form a single, large volume that equates to the sum of all drive’s capacity. JBOD does not provide data redundancy.
RAID 0 features “striping”, a configuration that utilizes 2 separate drives and writes “chunks” of data to each disk to minimize actuator movement to enhance performance. One such caveat is the lack of fault tolerance, if one HDD dies, all your data across the whole set of configured HDDs are gone. RAID 0 is only recommended if you do not plan to store any critical data onto the NAS.
RAID 1, otherwise known as “Disk Mirroring” replicates data to two or more disks within the same array. There is no increase in performance with this volume type, but if one drive fails the other operational HDDs will then be used until manual replacement of the failed drive. Once the RAID detects that the faulty drive has been replaced and is operational, it will then duplicate the contents of the secondary drive onto the new one. This ensure that the backed data is always readily available across all drives. Redundancy is improved with this configuration as more drive lots are filled.
A RAID 5 array requires a minimum of 3 or more drives with at least one drive reserved to rebuild the data on the array if any of the other drive dies. Data that is backed onto a RAID 5 setup is distributed evenly across all disks in such a configuration. In addition to data, parity information is also stored so that data can be recovered if one of the drive fails, but this process will be slower than RAID 1 as it needs to read the data on the reserved drive and rebuilt it from there.
Considered as an enhancement over to RAID 5 by many, RAID 6 utilizes block-level stripping and distributes dual-parity blocks on each disk within the array. A key feature of dual-parity block that puts RAID 6 above RAID 5 is the ability for operations to continue running on the NAS even if the stated failure drives of 2 drives have been reached. However, be advised that RAID 6 writes slower than RAID 5 as it needs to write to an additional parity block and requires a configuration of at least 4 HDDs to function.
RAID 10 is unique in the sense that this configuration combines two different RAID levels, RAID 1(mirroring) and RAID 0(striping) to achieve speed and redundancy. The implementation of RAID 10 requires the bare minimum of four HDDs to work.