We recently installed 45 HGST 8TB hard drives and we followed performance of the HGST drives for their data stacking stage to see how well they performed. This is the thing that we know along these lines.
The HGST 8TB drives are stacked with Helium. By HGST composing, the helium-filled environment lets the plates turn more easily as helium offers less resistance versus air. This suggests less power is required to turn the plates and they run cooler. It moreover suggests they can have more disks and examined/make heads with the HGST 8TB drive having 7 plates in a near space as various drives.
1. Our checking system reports that a drive has gone logged off in Pod 902.
2. Case 902 is flipped into upkeep mode. This places the Pod in read just mode, allowing any current to compose operations complete then simply allowing read operations.
3. Once the Pod is in upkeep mode, a datacenter expert uses the checking system to survey the issue planning to make sense of whether it is just a drive frustration or something more. Dependent upon the failure the drive can be restarted, yet habitually the drive is dead. That was the circumstance in Pod 902. Starting at this moment the drive is reserved for substitution.
4. Contingent upon workload and the season of-day, the failed drive is supplanted wherever from two or three minutes to a few hours afterward. The Pod continues working in read-only mode in the midst of that time.
5. To supplant the drive, the datacenter authority suspends new read operations then close down the Storage Pod.
6. The master finds the Storage Pod and slides it out of the rack. They open the Pod, locate the dead drive, certify it is the failed drive using the serial number, supplant the drive, and after that adjacent up the Pod. This consistently takes 3-5 minutes.
7. Once the Pod is back in the rack, it is restarted. The supplanted drive is then coordinated up into its RAID show. This can take wherever from two or three hours to a couple of days depending upon the degree of the drive. In the midst of this time, the Pod stays in upkeep mode, which implies it is scrutinized in a manner of speaking.
8. Once the new drive sync is done, the Pod is set again into full organization. This was the circumstance with Pod 902 and it continued stacking data ensuing to being in bolster (read) mode for around 49 hours.