Most recent computer operating systems (OSs) have been designed to work effectively with Advanced Format (AF) hard drives supporting 512 emulation (512e). AF 512e hard drives may deliver superior performance as compared to legacy 512 byte per sector drive models when properly configured with the host file system. In PC applications, the host file system is primarily determined by the OS installed on the computer.

Toshiba AF Whitepaper

More about Advanced Format can be found here.

Toshiba has presented the Discrete Track Recording Technology (DTR) in September 2007.
DTR is a new breakthrough recording technology that will increase the capacity of storage media used in PMR hard disk drives by up to 50 per cent. Therefore it will especially take the small form factor HDD’s, such as 1.8 inch and 2.5 inch drives, to a new level of enhanced capacity.
Small form factor HDDs are now found in such applications as mobile PCs, all kinds of portable multimedia players, digital video cameras, and also car navigation systems. This market has a strong demand for larger data capacities.

The DTR technique, as its name suggests, separates out parallel data tracks on HDD media by inserting gaps, also named “grooves”. This separation reduces signal interference between adjacent data tracks resulting in improved signal quality and allowing the pitch of the tracks to be shortened. The tracks can be made narrower, and more of them can be hosted onto the disk. This will enable recording densities of up to 516 megabits per mm2 (333 gigabits per square inch).

The technique is also applied to the servo pattern, some extra information added to the disk being an essential part of the read/write head positioning system.

DTR technology is based on research of Japan's New Energy and Industrial Technology Development Organization (NEDO).

Toshiba expects to lead the industry in mass production of HDD integrating DTR technology. The current plan is to put DTR drives into mass-production during 2009.

Traditionally, Toshiba HDD have ball bearing spindle motors. These ball bearings are metal balls that are lubricated by a thin layer of grease. Imperfections in the roundness of the bearings and within the raceways (in which the bearings reside) set up random vibrations that can cause problems (increased "Non-Repeatable Run Out" - NRRO) for the servo system. Additionally, the ball bearings start to wear with continuous use, even more so when subject to excessive shock. A symptom of this wear is the spindle motor becoming noisier and read/write performance decreasing.

A solution to these issues is to use spindle motors that use a fluid dynamic bearing (FDB). FDB motors have lubricant oil rather than metal ball bearings to separate the rotor and stator.
Advantages of using a FDB motor:

• Greatly reduce NRRO to improve servo performance. This will allow for higher areal density.
• Improved reliability by not having any metal ball bearing.
• Quieter due to not having metal ball bearings.
• Increase in non-operating shock performance.

Notebooks have been developed to increase mobility. A small form factor hard disk drive such as 2.5" and 1.8" HDD is one of it’s main components and it is therefore exposed to the same risks as the notebook itself: Shock, vibration and – even more serious – the impact after a drop.

One of the most harmful things that could happen to your hard disk drive is a so called "head disk interference" (HDI), which means nothing else than a collision of the most sensitive parts of a HDD, the read/write heads and the surface of the storage media. Through the impact, not only the heads can be damaged, but vital data stored on the hard disk can also be damaged or lost. HDI are mostly caused by shock impacts, e.g. when a laptop is dropped.

Toshiba Storage Products Division has developed an effective technology to prevent damages caused by an impact after a free fall. The principle is as simple as it is effective: When a possible drop acceleration is measured, the read/write heads are retracted from their location above the media and locked in a secure position. In other words: The shock robustness of the drive is significantly increased.

Obviously, the technology is more complicated. A three axis sensor, a so called "Low-Power Linear Accelerometer", measures the acceleration magnitude, which in normal conditions is approx. 1 G. When the hard disk drive is falling, the acceleration magnitude changes to approx. 0 G. The sensor detects this transition phase from 1G to 0G and retracts the read/write heads from the media preventing HDI.

The accelerometer is so fast, that a fall from as little as 10 cm (4 inches) is detected and the read/write heads are secured. In other words, the whole process of measuring the change in G force, retracting and securing the heads lasts less than 150 milliseconds.

Toshiba hard disk drives with Free Fall Sensor technology are the first choice for truly mobile storage devices.

HAMR is an acronym for Heat-Assisted Magnetic Recording.

It describes a future HDD recording technology that allows even higher areal densities on Hard Disk Drive media than achieved with Perpendicular Magnetic Recording (PMR).

With HAMR data is magnetically recorded on special high-stability storage media after the precise spot where data bits are being recorded has been heated up using laser thermal assistance. When heated, the medium becomes easier to write, and the rapid subsequent cooling stabilizes the written data.
These media materials can store single bits in a much smaller area without being limited by the "Superparamagnetic Effect". This is the same effect that limits the areal densities of platters used in conventional hard disk drives recorded with LMR (Longitudinal Magnetic Recording).
The only catch being that they must be heated to apply the changes in magnetic orientation.

HAMR will not appear in general HDD products before 2010 and 50 Terabits per square inch areal densities may be possible in the 2020 or later area. The transition from PMR to HAMR may well have started in 2010.

The variety of products using hard disk storage is vast and growing. More and more consumer products require an increasing storage capacity to meet modern demands. The range of products supported by Toshiba storage technology extends to meet this demand. Toshiba was the world’s first to launch Hard Disk Drives (HDDs) with Low Insertion Force (LIF) connectors. The new LIF SATA interface presents a 10% smaller footprint product when compared to standard 1.8-inch drives using micro SATA connectors. This smaller footprint connector ensures that Toshiba products are best placed to meet the demands of major manufacturers of small mobile devices such as tablet PCs, digital camcorders and media players.

                    ZIFF HDD                                                   LIF HDD

In today’s modern computer systems, support for multi-tasking and multi-user environments is a fundamental requirement.  We often have multiple activities running, such as watching movies, playing music, checking emails, surfing the internet and sometimes the sharing of common system resources with other users.  

Native Command Queuing (NCQ) is a technology designed to increase performance and reliability of SATA hard disk drives (HDD) under certain workload conditions by allowing the HDD to intelligently optimise the order in which the received read and write commands are executed. This can improve the HDD command processing efficiency, thereby reducing the mechanical workload and increasing the overall performance of the drive. Without NCQ, the HDD would complete each read and write command in the order they were received, which decreases the efficiency and performance.

The main benefits of NCQ:

• Ideal for multi-tasking and multi-user environments
• Works in all systems where host controllers support the SATA NCQ feature including desktop PCs, workstations, digital media content servers, entry servers, as well as high performance PCs and notebook systems
• Provides 100% backward compatibility with non-NCQ supporting systems
• Allows the reordering of commands by the storage device to increase the performance efficiency of its data transfers
• Improves seek time performance for HDDs and allows solid state drives to access the stored command queue in order to boost performance

NCQ is often used to increase performance in heavy workload servers and high-performance workstations. It is also used to optimize PC performance in operations such as file copying and system booting.

Conventional hard disk drives using "Longitudinal Magnetic Recording" (LMR) store data on a magnetic disk as microscopic magnet bits aligned in plane. Although advances in magnetic coatings continue to improve data recording densities on HDD, the magnetic bits repulse each other due to in-plane alignment.

Squeezing more bits on to a disk will eventually reach a point where crowding degrades recorded bit quality because the bits start to influence each other (magnetic coupling). In worst case they begin to flip over: The affected bits loose orientation and the information is lost. This is called the "Superparamagnetic Effect". This problem placed fast-approaching limits on storage capacities.

The solution to this problem is called "Perpendicular Magnetic Recording" (PMR) where the magnetic bits are not aligned in plane but orientated vertically. By standing the magnetic bits on end, perpendicular recording reinforces magnetic coupling between neighbouring bits, achieving stable higher recording densities and improved storage capacity. By using PMR the capacity of hard disk drives can be increased by up to 10 times. Furthermore due to the increased data density the data transfer rate is also increasing compared with HDD’s using the LMR technique.

One of the technical challenges when developing PMR was the fact that the physical characteristics of the PMR media require a much narrower gap between read/write head and media in order to be able to read and write data.

Professor Shun'ichi Iwasaki from Tohoku Institute of Technology in Japan initially determined in 1976 that it was possible to increase areal density of data media by organising magnetic bits vertically to the rotational direction of the media rather than horizontally.

Anti-Rotational Vibration technology has been developed by Toshiba to counteract the impact of system vibration on Toshiba hard disk drives (HDD).

Rotational Vibration Sensors (RVS) enable Toshiba HDDs to operate more efficiently in environments susceptible to high rotational vibration such as servers or storage arrays. When a HDD is exposed to rotational vibration, the performance can be degraded. With a few drives in close proximity, such as in a storage array, each drive is vibrating as a natural result of head positioning on the rotating media. The cumulative effect of these drive operations sometimes creates harmonics that induce sharply higher vibrations. As activities of HDDs affect the whole array, the operation of each drive in the system can progressively be disturbed resulting a significant drop in data transfer rate. Sensitivity to these vibrations grows as drive data storage density continues to increase, resulting sometimes with smaller vibrations knocking read and write heads off track.

The new RVS technology used in Toshiba HDDs senses the external rotational vibration and uses a feedback system to produce a counter movement in HDD read and write heads. RVS enables the drive to compensate for any rotational vibration that occurs from the drive itself, or outside of the drive (cooling fans, poorer-quality chassis etc), and continues to read and write whilst maintaining its high performance.
 

The information on a Hard Disk Drive, known as HDD, is often more precious than the device itself. A company’s data is one of its most valuable assets. Unauthorized data exposure can occur anywhere – from the one-man IT department to the nest-managed data centre. Software-based solutions cannot provide the security required for a totally reliable protection. The new hardware-based solution called SED is now used to establish a strong digital identity, taking security to a higher level.

Toshiba's 2.5-inch self-encrypting drives (SEDs) provide both advanced hardware encryption and strong access authentication to help IT departments cost-effectively deploy strong security without interrupting business flow or impacting application performance. Designed to the Trusted Computing Group “Opal SSC” specification, Toshiba's SED models benefit from the broad industry support available for industry-standard TCG security features and capabilities.

SED encryption processes are transparent to applications and operating systems. Unlike software encryption, which is dependent on CPU performance and system memory capacity, the SED's hardware encryption occurs inside the HDD at full storage I/O speeds, ensuring that users will not see a reduction in application performance.

Data Invalitation Technology - A Step Forward in Data Security

Built-in hardware encryption within the HDD’s controller electronics offers performance, lower overall cost and security benefits beyond those available with software encryption. Toshiba's unique and strong, government-grade AES-256 encryption is certified by the US National Institute of Standards and Technology (NIST) through its Cryptographic Algorithm Validation Program (CAVP). Due to the encryption process happening inside the HDD, stored data remains safe from all kinds of attacks used to compromise software encryption.

Due to the need of increased bandwidth and flexibility, SAS has been introduced to replace the parallel SCSI interface that first appeared in the 80s in data centers and workstations. Compared to SCSI, SAS has several advantages. It can operate at more than 10,000 RPM and is more reliable. SAS drives are typically utilized in server and high-end workstation in environments where speed and I/0 frequency is very important. In addition, with SAS drives, users have full error detection and error correction during the read and write process. It also provides data protection while data is in flight, which ensures the integrity of data during reading and writing. Enterprise Class SAS drives are preferred in online and transactional applications due to the exceptional performance and reliability.

With a seemingly exponential continued growth of data, the management of that data and the demands placed by today’s enterprise computing environments and applications, it becomes increasingly more vital to ensure accessibility, availability, performance and reliability criterion are met 24x7. Running  in parallel with these expectations, advances in storage technology and techniques using Solid State Drives (SSD) has meant that the quality of NAND based flash memory has now reached position, where it can provide mission-critical data reliability alongside data rates offering considerable improvements over traditional rotating HDD.  SSD effectively places semi-conductor memory behind a memory controller and interface electronics allowing the flash storage device to emulate the commands and operations of a HDD.

Toshiba’s first generation eSSD uses 32nm Enterprise grade Single-Level Cell (SLC) NAND flash memory and 6Gb/s Serial SCSI interface.

eSSD is predicted to replace the traditional mechanical spindle based technology in the majority of high-end performance led applications. It makes an excellent fit for mission critical applications where reliability, performance and long term expectations are desired.

In Enterprise Solid State Disks (eSSD), a Single-Level Cell (SLC) is a memory element capable of storing data in individual memory cells.

SLC accesses and writes to the eSSD by using „simpler“ control logic with 1 bit versus the 2 bits used by Multi-level Cell (MLC) and has the advantage of lower power consumption.  In addition, the program operations in SLC chips last 100,000 cycles, ten times longer than MLC. Due to the very fast transfer rate and high reliability, SLC memory is used in high-performance SSDs.

This is the maximum number of Program/Erase cycles that a cell is designed to achieve before reaching end of life and cell structure breakdown causing loss of data retention and reliability.

This is the time period that the stored cell data is expected to remain recoverable and uncorrupted during the life of the storage device. The cell structure degrades as it approaches maximum endurance or in other words the closer the cell Program/Erase (PE) count gets to the value representing total life expectancy.” When looking at long-term storage of data under power-off condition (shelf life / archiving), the period can be as low as 3 months if a cell is at 100% endurance due to the greater potential for the breakdown of the cell and loss of stored value.

This is essentially excess capacity allocated to the storage device to allow for cells reaching 100% endurance (life expectancy reached) within the period of warranted use of the storage device. Excess capacity is brought on-line and allocated to offset the loss of usable storage cells. Generally the acceptable figure runs between 25-28% of purchased capacity point.

Each memory cell has a finite life, their structure breaks down and they wear out! To ensure that each cell is subjected to equal wear and tear a Progam /Erase (PE) count is maintained at block (sector) level. Once the count has reached a pre-determined threshold, the collective cell data content is re-assigned to a younger cell collective / sector. Two types of implementation exist:
Static Wear levelling is an algorithm aimed at stored data that rarely changes such as OS files. These cells would remain underutilised if left alone, so the data content is re-assigned to more mature cells freeing up the younger cells for more frequent change data storage.
Dynamic Wear levelling is the general purpose algorithm that is executed every time the data in the buffer is flushed and written to flash. The dynamic wear algorithm guarantees that data program and erase cycles will be evenly distributed throughout all the blocks within the NAND

NAND devices store one bit of information per memory cell. A SLC drive has faster performance, lower power consumption but lower capacity.

NAND devices can store more than one bit of information per memory cell by choosing between multiple levels of electrical charge to apply to the floating gates of its cells. This technology enables higher density storage in a small form factor and more cost-effective storage per gigabyte

Class process technology or die size relates to current CMOS semiconductor NAND device fabrication technique. This equates to the transistor gate pitch or in other words the transistor packing density making up the memory cells (around 2 million).

Technology advancement is estimated as follows:
32 nm — 2010
22 nm — approx. 2011
16 nm — approx. 2013
11 nm — approx. 2015

Provide power backup for a limited period of time to SDRAM cache memory in order to preserve the data contents and protect from corruption or loss. The Power Loss Protection (PLP) is designed to last 55ms with Toshiba eSSD design to allow the cache contents to be flushed and written to NAND flash.