I have a feeling we might look back in a few years in the same about solid state drives. Yes they have some benefits, but also some drawbacks. In the last week two SSDs have come in for recovery. An early 16GB mini PCI-E device found in a Dell mini using an STEC controller and a more modern 64GB OCZ Vertex 3. Bad news for the OCZ customer is that the Vertex 3 uses a Sandforce controller, meaning the data is encrypted on the Flash NAND chips. The OCZ Vertex 4 have moved back to using Indilinx controllers which hopefully will not encrypt the data which should allow for greater possibilities when it comes to data recovery.

The process of recovering data from SSDs is a challenging one, and consequently the pricing for the recovery of data from these devices can be somewhat high than that of hard disk drives. Do back up your data, particularly if you have a Sandforce controller.

In the example below a Seagate Barracuda 7200.10  drive is used.

Key to Illustration Inside a Hard Drive:

1. Air filter.
2. Platter spacers.
3. Permanent magnets.
4. Platter landing zone.
5. Platter motor.
6. Data platter.
7. Read/Write heads’ pre-amplifier circuit (in this case mounted on the side of the arm assembly).
8. Read/write heads.
9. Connections from the read/write heads through to the printed circuit board at the back.

Office hours:
Monday, Tuesday, Wednesday: 8.00am – 12.30pm
Thursday: Out of Office
Friday: 2pm – 7pm
Saturday & Sunday: 10am – 5pm

Please note an out of hours disk diagnosis service is still available. For all contact details please visit the main contact page.

As the cost per gigabyte decreases they are becoming increasingly popular as storage devices where hard disk drives would be used. The most significant difference between flash NAND technology and hard disk technology, is that the former does not have any mechanical components.

Here we have two images of flash based media with their outer packaging removed:

Inside a solid state disk - what you would expect to find

A flash based memory stick. Highlighted in red is the controller chip

The SSD and memory sticks that Cheadle Data Recovery receives in for recovery usually relate to a failure of the controller chip. This is the component that allows the storage device to interact with the computer and control the way data is distributed through the flash nand chips. The controller converts requests for logical sectors (LBA values) into the physical locations on the actual flash memory chips. It also contains a wear levelling algorithm which extends the working life time of the device.

When a controller chip fails one the most challenging aspects is calculating  where data is spread through the flash chip. Data is not recorded in a simple linear fashion. The controller chip is designed to increase the performance which means spreading data through multiple locations to allow, rather than having required data in sequential blocks. As such recovery of data from these devices can be extremely challenging.

For further reading please see this thorough introduction to Flash NAND media.

Given how obvious the fault can be many users assume that recovery of the fault will be equally as obvious – find a PCB, put it on the HDD and “hey presto” a working HDD. Sadly on nearly all modern HDDs this is not the case, as most PCBs have unique data on them. This unique data (adaptive data) needs to match content in the firmware which is based on the platter surface. This means that when we find a suitable we need to remove component from the failed patient PCB or recreate the content using data found in the firmware service area.

Western Digital HDDs can be particularly challenging to find donor parts for. The reason for this is the frequent revision changes on critical parameters. The example below denotes just some of the requirements when considering to work on this hard disk drive. There is another, which is not listed on the HDD itself. This relates to the ROM / ATA module. There is no way of knowing this just by looking at the drive. As you can imagine this can be very frustrating! Consequently it is wise to keep a large stock of donor drives, which we do.

For all those trying to find donor parts a mention must go to Access Computer Parts / Fred Farzan based in California. Everyone that has worked in the data recovery industry will be familiar with Mr. Farzan. If there’s a rare hard disk drive you need that you can’t find any where else, he’ll have it.

500GB on a single platter for a laptop 2.5 inch hard disk drive.  Quite how shock resistant it might prove is another matter. “400G” doesn’t mean much to most people. Given that many of the hard disk drives we see through with mechanical failure (usually making a clicking or a buzzing noise) and are a result of physical shock, as such any improvement would be good.

“A new report from IDC has confirmed the fact that worldwide shipments of hard-disk drive units experienced a year-over-year decline of 4.5 percent in 2011.”

 

“According to the IDC forecast, the HDD industry will record year-over-year unit shipment growth of 7.7 percent in 2012 and a compound annual growth rate (CAGR) of 9.6 percent from 2011 to 2016.”

Gosh, that caused a yawn. The point is HDD sales took a dent in 2011. They seem to be recovering, people still want to buy them, and prices have finally settled. USB3 devices are becoming more common place, which is a relief, as transferring hundreds of gigabytes of data on a USB2 connection can be painful. From the systems we have here it still looks as if USB3 struggles when transferring lots of small files (high I/O requests). Throughput on large files is good, but safe to say SATA connections still rule the roost for transfer speeds in the consumer market.

In recent years Hitachi have been making some good hard disk drives. We haven’t actually seen many of their desktop HDDs in, they have tended to be relatively reliable. The Hitachi firmware architecture was pleasant to work with the only exception of problems with the NVRAM and having to find an awkward donor part. The same could be said for Samsung too.

So just left with Seagate and Western Digital. From a data recovery standpoint both manufacturers make ‘awkward’ HDDs to perform data recovery on. The Seagate firmware F3 architecture is not as simple to manipulate as on older series (Barracuda 7200.10 and earlier) and it’s not possible to initialise the hard disk from a single head. The WD ROYL series can be problematic also, not least the issues associated with their integrated ROM

Ohh dear, I expect the story did not have a happy ending for this hard disk drive

All of this won’t mean much to most people. What will is our favourite bit of IBM/Hitachi hard disk history with the infamous Deskstar GXP75. I still see a few ‘Deathstars’ come in for recovery. I’m amazed they’ve lasted so long. When they do come in they have all manner of faults. I won’t miss these.

It’s not just OEMs, computer shops and data recovery companies that running short of hard disk storage space due to the flooding in Thailand last year. Now it seems that the Large Hadron Collider (and its associated computer network) have run short of disk space.

Crunching the deluge of data coming out of the LHC experiment relies on a network of computers called the International Science Grid that stores, shares and processes the information.

“But we’re crying out for storage, and the floods in Thailand didn’t help. It’s compromising our experiment,” he explained. “We have seven petabytes of storage and it’s not enough.”

The CERN experiment’s hardware emits a raw flow of 50 million petabytes a year. The majority of that data is discarded, reducing the wedge to 15PB, and then split between computers on the International Science Grid.

All in all, that’s a lot of data to contend with.

More storage, greater, bigger, faster! (gasp) Seagate hope to extend the capacity of magnetic storage media (hard disk drives) even further using heat-assisted magnetic recording (HAMR), which looks set to replace perpendicular magnetic recording (PMR) in the future.

It hopes to be the first manufacturer to reach the storage density of 1TB (1 trillion bits) per square inch. With the use of this technology Seagate hopes to reach sizes up to 60TB. The current limit on PMR drives is about 3TB

“Hard disk drive innovations like HAMR will be a key enabler of the development of even more data-intense applications in the future, extending the ways businesses and consumers worldwide use, manage and store digital content,” stated  Mark Re, the senior vice president of heads and media research and development at Seagate.

What this quite means for the data recovery industry – increased capacities mean more data, and typically more time processing hard disk drives. Challenging which ever way you look at it.