Since their inception, HDDs employed the longitudinal mode of recording. However, 2005–2007 saw the long-awaited introduction of a new mode of recording: perpendicular recording. In perpendicular recording the magnetization axis is orthogonal to the disk surface rather than lying in plane. Perpendicular recording has the virtue of writing very sharp transitions into a relatively thick, high-coercivity medium [2]. The introduction of perpendicular recording, in particular, has reinvigorated HDD technology. The areal-density growth curve which had been flagging has taken an upward tick to around 50% per annum in very recent products (Fig. 1).
However, there is one fundamental limit that even perpendicular recording cannot avoid and that is the thermal or ‘superparamagnetic’ limit [3,4]. In current systems, the noise of the transitions is not far from the lower limit that occurs when the closest to the ideal position. For such a ‘grain-limited transition’, further large reductions in noise can only be achieved by reducing the grain size. With reasonable expectations for head fields, media thickness, improved media switching properties, and tighter grain-size distributions, we might expect grains of 6 or 7 nm diameter to be writable and thermally stable. With some help from signal processing, the limiting areal density is estimated to be approximately 1 Terabit/sq.in. for conventional recording [5,6]. Beyond 1 Terabit/sq.in., radically new approaches will be required including thermally-assisted magnetic recording (TAR), where a laser is used to heat a tiny spot on the recording medium, and patterned media where one bit is assigned to each predefined magnetic island. There are formidable challenges in implementing either of these options from the perspective both of perfecting the required technologies as well as creating largescale low-cost manufacturing processes. Beyond the next one or two generations of products, the frantic pace of areal-density growth is likely to slow significantly as we approach the 1 Terabit/ sq.in. regime.
Since their inception, HDDs employed the longitudinal mode of recording. However, 2005–2007 saw the long-awaited introduction of a new mode of recording: perpendicular recording. In perpendicular recording the magnetization axis is orthogonal to the disk surface rather than lying in plane. Perpendicular recording has the virtue of writing very sharp transitions into a relatively thick, high-coercivity medium [2]. The introduction of perpendicular recording, in particular, has reinvigorated HDD technology. The areal-density growth curve which had been flagging has taken an upward tick to around 50% per annum in very recent products (Fig. 1).
However, there is one fundamental limit that even perpendicular recording cannot avoid and that is the thermal or ‘superparamagnetic’ limit [3,4]. In current systems, the noise of the transitions is not far from the lower limit that occurs when the closest to the ideal position. For such a ‘grain-limited transition’, further large reductions in noise can only be achieved by reducing the grain size. With reasonable expectations for head fields, media thickness, improved media switching properties, and tighter grain-size distributions, we might expect grains of 6 or 7 nm diameter to be writable and thermally stable. With some help from signal processing, the limiting areal density is estimated to be approximately 1 Terabit/sq.in. for conventional recording [5,6]. Beyond 1 Terabit/sq.in., radically new approaches will be required including thermally-assisted magnetic recording (TAR), where a laser is used to heat a tiny spot on the recording medium, and patterned media where one bit is assigned to each predefined magnetic island. There are formidable challenges in implementing either of these options from the perspective both of perfecting the required technologies as well as creating largescale low-cost manufacturing processes. Beyond the next one or two generations of products, the frantic pace of areal-density growth is likely to slow significantly as we approach the 1 Terabit/ sq.in. regime.
การแปล กรุณารอสักครู่..
