Save time using software defined storage
Software defined storage (SDS) has many advantages in terms of cost and removing some of the complexity, at least on the surface, of managing the sheer amount of data in a mature organisation. It can also save CTOs time in a number of ways:
- Deployment: re-tooling for a new storage system can be a considerable undertaking because in a traditional system the software and hardware are interlinked. Even single users can suffer from this as an individual external disk for an Apple computer won’t be read by a PC and vice versa. By decoupling the hardware and software, SDS makes re-tooling much more straightforward as the hardware can be replaced and the software will simply find it. This takes away a lot of the pain of upgrades.
- It’s a little “chicken and egg” but as SDS depends mostly on x86-based architectures, it takes away a lot of the research and agonising that might have happened before as a result of changing a storage implementation. The fact that it simply works is a major step forward.
- Automation: a related point is that the standardisation of SSD allows for the configuration and indeed the policy of an implementation to happen at the API or CLI level. And anything that happens automatically is going to take less time than something that needs more thinking through.
- Pooling: SDS does not know the difference between media, so it is possible to pool data between unlike disks and to use them as one. This means less time swapping out older hardware to substitute newer items.
- Finally a good SDS system will transparently move and migrate data. IT directors and professionals spend a lot of time after rolling out an upgrade just getting back to where they were before; the software’s ability to find its own way is a considerable boost for such personnel.
It would be wrong to suggest all software defined storage is equal or that every iteration will work identically regardless of expert implementation. However, SDS offers the CTO an opportunity to save a lot of time that will be better spent on strategy than on remaking a system every time there is an upgrade.