Reliability
Reliability Reinvented!
Enabled by its grid architecture and unique RAID-X protection scheme, the Nextra system provides unprecedented reliability. These qualities immunize the Nextra system against any single failure and reduce the risk of double failure significantly below the standard.
Nextra achieves its unmatched reliability through various means:
- Truly no single point of failure – The Nextra system can survive any single failure
without affecting host I/O
- 15-minute rebuild time for 500 GB drives – The Nextra system is designed to minimize the possibility
of double disk failure
- Rebuild of real data only – The Nextra system performs rebuild on only that data which has been allocated to volumes and actually written
- Self-healing upon module failure – The Nextra system self-heals even after module failure, initiating a rebuild process and returning to full redundancy
- Consistent performance through any failure – The Nextra system maintains the same high performance level through any failure
No single point of failure
The Nextra storage system is a true no single-point-of-failure system. All system components
are protected in an active-active N+1 scheme, ensuring both high reliability and excellent
performance. The Nextra system can survive any single failure
without any influence on host I/O. Every disk, module,
switch and UPS is redundant and protected. Each component is
hot-replaceable: the entire system
can be replaced without ever being shut down.
15-minute rebuild time
The Nextra system uses innovative RAID-X design, in which each disk is split into small pieces, and each piece is mirrored on a different disk. As a result, when a disk does fail, all disks in the system participate in the rebuild.
After a failure of a 500 GB drive on a fully utilized system, the system is exposed to double disk failure for a rebuild
time of just 15 minutes. This time significantly
reduces (by orders of magnitude) the risk of data loss
from double disk failure, especially
in comparison with other storage systems.
Rebuilds real data only
The Nextra system performs rebuild on only that data which has been allocated to volumes and, of the allocated data, only on data that has actually been written. Given that in most scenarios not all capacity is allocated and not all allocated capacity is used, Nextra's rebuild time is, in practice, much less than 15 minutes. Other storage systems typically perform disk rebuild at the block level, rebuilding the failed disk completely and taking the full length of time this requires.
Self-healing upon module failure
The Nextra system employs self-healing even after module failure by initiating a rebuild process and returning to full redundancy. The failed module is replaced only after redundancy has resumed, protecting the system from technician errors.
