Reference
Glossary of data recovery
Every diagnosis contains words that nobody outside this field needs. Here they are in comprehensible form — so that you can read a quotation without having it translated.
- Adaptive values
- Individual calibration values determined at the factory for one particular drive and stored on its circuit board. They are the reason why a board swap does not work without transferring these values.
- Bad block / bad sector
- A location that can no longer be read reliably. Drives replace such locations internally with spares — nothing is repaired in the process, and the data sitting there is usually lost.
- Chip-off
- Desoldering the memory chip in order to read it out away from the board. Necessary when the electronics are destroyed. On encrypting devices the method yields only ciphertext and therefore does not lead to the goal.
- Clean room
- An environment with a heavily reduced number of particles in the air. Necessary in order to open hard drives: the head floats a few nanometres above the platter, and a speck of dust is many times that size.
- CMR and SMR
- Two writing methods on hard drives. With SMR the tracks overlap like roof tiles — that increases capacity but makes every modification laborious and the firmware considerably more complicated. For data recovery SMR is the harder case.
- Controller
- The control chip in an SSD, memory card or USB stick. It manages where which data physically sits, distributes write operations and, on many models, encrypts. If it fails, the memory is usually intact but out of reach.
- Degraded
- The state of a RAID array that is missing a member. It keeps running but no longer has any reserve — the next error gets through.
- Donor drive
- An identical drive from which spare parts are taken. “Identical” here means more than the same model: head allocation, preamplifier, firmware generation and production series all have to fit together.
- ECC (error correction)
- Additional check information with which flash memory corrects individual flipped bits. When reading memory chips directly, this correction has to be reproduced externally — otherwise the content makes no sense.
- File system
- The administrative structure that turns blocks into files: names, folders, timestamps and the information about which blocks belong to which file. Common representatives: NTFS, exFAT, APFS, HFS+, ext4, Btrfs, XFS, ZFS, ReFS.
- Firmware
- The software inside the storage device itself. It is the reason why a mechanically perfect drive can nevertheless be “not detected”.
- Garbage collection
- The clean-up process in SSDs and memory cards: blocks marked as invalid are emptied in the background so that they can be written to again. It runs on its own as soon as the device has power.
- Head crash
- The read/write head touches the magnetic platter and abrades coating. The debris creates new particles that cause further damage — a self-reinforcing process. Recognisable by grinding or scratching noises.
- Head stack
- The assembly containing all the read/write heads of a hard drive together with the preamplifier. It is replaced as a whole during a head replacement.
- Image
- A sector-by-sector copy of a storage device, including the areas where no files are located. The basis of every serious data recovery: all further work runs on the image so that the original is not written to.
- LBA (logical block address)
- The consecutively numbered address with which an operating system addresses a block. Where that block physically sits is known only to the storage device itself.
- Micro-soldering
- Work on individual components on the circuit board under a microscope. The route for electronic damage to SSDs, memory cards, smartphones and soldered notebook storage.
- Monolith
- A design in which the controller and memory sit in a single potted package — the rule for microSD cards. There is nothing to desolder; access has to be established via exposed contact points.
- Parity
- The check information in a RAID array from which the content of a missing drive can be calculated. During a rebuild it is written anew — even when the basis for doing so is already incomplete.
- Preamplifier
- A tiny component on the head stack that amplifies the extremely weak signals from the heads. If it does not match the drive, the heads are driven incorrectly — which is why the model number alone is not enough for a head replacement.
- Raw data recovery
- Searching for files by their characteristic opening bytes when the file system can no longer be reconstructed. Returns content, but no names, folders or timestamps.
- Rebuild
- Restoring a RAID array after replacing a drive. It reads all remaining members under full load for days and is thus the riskiest measure at exactly the moment when there is no reserve left.
- Service area
- An area invisible to the operating system in which a hard drive stores its own administration — firmware modules, defect lists and calibration values. If it is damaged, the drive reports itself incorrectly or not at all.
- SMART
- Self-monitoring of the storage device. Counts, among other things, reallocated sectors, read errors, operating hours and temperature. Useful for prevention, but not a prediction.
- Stripe / block size
- The amount of data written contiguously to one drive in a RAID array before the next one takes its turn. One of four values that have to be right for a reconstruction.
- Translator
- The internal translation table of a storage device between logical address and physical position. If it is damaged, the device reports an incorrect capacity or supplies no data at all — even though the memory is intact.
- TRIM
- The command with which the operating system tells an SSD which areas are no longer needed. After that the drive returns nothing but zeros there. The main reason why deleted files on SSDs are usually not recoverable.
- Wear levelling
- Distribution of write operations across the entire flash memory so that individual cells do not wear out prematurely. One reason why a file physically sits in a completely different place than the operating system assumes.
A specific case?
We will take a look at your storage device.
The Economy diagnosis is free of charge. You find out what is possible before anything costs money — and what it would cost is put in writing beforehand.
Wasserweg 8–10 · 60594 Frankfurt am Main