RETRODATA

Any housing, casing or dedicated storage system that contains more than one hard drive working in unison with all the other drives with the purpose of storing and sharing data is almost certainly going to be a RAID storage device.

Most home users and smaller businesses will have some sort of NAS (Network Attached Storage) device with one or more hard drives. Unfortunately, most of the low-end units use a configuration of RAID 0. However, RAID 0 is an oxymoron; RAID stands for “Redundant Array of Independent Disks” – with “redundant” meaning that even if a single drive fails, the data will still be accessible. Not so with this one. A failed RAID 0 array (even with a single drive failure) will prevent all access to the data and the services of a RAID data recovery company will be required. These low-end NAS systems almost invariably have little or no active cooling, and should be avoided where possible.
Continue reading →

MO covers a broad spectrum of optical discs (which include the now-defunct MiniDisc) that can be written to and read from – albeit rather sluggishly in comparison with modern storage technology.

Magneto Optical discs are no longer commonly used, other than for archiving historic data, and they have become largely redundant, having been replaced by alternative, less-costly storage such as DVD.

There remains a strong need for data recovery or extraction from these discs; for example, many hospitals use them for patient data, stored in the DICOM format. Due to the proprietary nature of the discs, normally extraction has to be performed from the same manufacturer (and often the same model) of the MO drive itself. A prime candidate is the Pioneer DC-502A – at one point apparently the standard for DICOM data storage.

RAID arrays form by far the most popular modern method of storing data.

Originally designed with two factors in mind; more reliable data storage (through redundancy – which does not apply to RAID 0 or spanned arrays) and an increase in throughput performance, i.e. far quicker operation than a single drive.

There are dozens of different RAID configurations, known as levels. The type of application, the storage requirement, scaleability, expandability and performance are all factors which will dictate the best RAID level to implement. RAID 30 for high-speed video editing, RAID 1 for metadata and log storage, and RAID 5 (possibly the most common level configured) for all-round use.

Most “intelligent” computer components have firmware. Firmware can be though of as a translator between the operating system and the hardware component itself, and also as a software program that controls certain parameters of the device.

Hard drives all contain firmware, which is responsible not only for “decoding” the interface between operating system and data stored on the drive, but it’s also capable of controlling slider fly height (the height at which the read/write heads, embedded into the slider, fly above the surface platters) as well as drive operating volume, error correction, and a host of other hard drive operations.

Usually, the firmware on a hard drive requires updating only when there is a specific flaw. Seagate experienced this early in 2009, with their 7200.11 series of hard drives. They did issue a firmware update that rectified the problem – but not before many people had spent money on expensive data recovery.

Computer systems contain firmware – usually on the motherboard – but again, this is not typically updated by users with any regularity, unless an issue is discovered following its release. That said, motherboard manufacturers do release firmware updates from time to time.

NAS storage devices are the devices we most often encounter with firmware (or “software”) update problems. Manufacturers of these devices appear to release numerous firmware updates. We imagine that one of the main reasons for this is that they need to be compatible with as many software packages as possible, as well as with all manner of hardware devices likely to be attached to the NAS.

However, firmware or software updates with NAS are fraught with danger, and it is strongly advised to make a secure backup of all the data on the NAS before updating the firmware.

Firmware failures on NAS devices can cause loss of access, corruption to the RAID or JBOD parameters, file system corruption and even reinitialisation of the disk or array.

Formatting a drive simply means using an operating system to place a file system on a storage device so that it can be transparently written to and read from.

It is unusual to have to perform a re-format; typical scenarious could involve suspecting there is a virus on the system that cannot be removed, or the storage medium is being prepared for a different operating system. Lastly, it happens in the unfortunate system when a storage device (such as a RAID array) has been corrupted or failed by logical means, and the only way forward is seen as re-formatting the array.

Most modern RAID controllers and NAS boxes run a low-level media check on all drives, and actually perform a full format, effectively zeroing the drive and permanently and irrevocably destroying any data. (This is one of the reasons a format takes so long on these storage devices.)

Coming soon…

Apple Xsan is an incredibly high-performance, fibre-channel, clustered file system configured as RAID within a storage area network (SAN).

The storage hardware consists of data servers (usually individually configured with RAID 3 or RAID 5, and then striped with each other) and separate Metadata servers which, unsurprisingly, containg the Metadata for the user data stored on the data servers. The Metadata typically makes up around 10% of the total data capacity.

Data recovery of Xsan is incredibly complex, and a task that only a handful of companies in the world are able to carry out, with Retrodata being one of them.