Enbedded eMMC, UFS, eMCP and PCIe-based SSD PCIe NVMe
SSD are traditionally used in industrial applications such as factory
automation, medical, and transportation. These ICs design relatively new in the
industrial segment and widely used today. In addition to having a removable M.2
form factor, the BGA form factor offers a highly ruggedized solution for
industrial applications, providing up to 1TB of storage space. Systems
requiring greater than 1TB can utilize multiples of these devices to take
advantage of the 16mm x 20mm package size.
Automotive MCP not only
need memory configurations with the right mix of density, power, performance,
temperature, reliability and cost, but they also require solutions in a very
small footprint. The problem
is not just in storing raw data, though. For businesses to stay competitive,
they need to quickly and cost-effectively access and process all that data for
business insights, research, artificial intelligence (AI), and other uses. Both
memory and storage are required to enable this level of processing, and
companies struggle to balance high costs against limited capacities and
performance constraints.
The challenge is even more daunting
because different types of memory and storage are required for different
workloads. Furthermore, multiple technologies might be used together to achieve
the optimal tradeoff in cost versus performance. MCPs are optimal
solutions for V2X communications. MCPs solutions have been limited by density, performance, and cost.
This limitation has been felt across all kinds of organizations, from retail to
government to healthcare to finance. For example, cloud service providers
(CSPs) can struggle to meet service level agreements (SLAs) as data loads
increase. Financial-services companies can run up against capacity and
performance limits for rapidly processing high volumes of transactions. And
enterprise businesses can’t keep pace with in-memory analytics needs stemming
from customer, inventory, social media, and Internet of Things (IoT) data —
primarily due to the high costs and limited capacity of dynamic RAM (DRAM).
To manage
data efficiently and effectively, businesses need to determine which
infrastructure components best match their needs and budgets. That’s no easy
task because each technology in the hierarchy has its strengths and weaknesses:
- DRAM is
great for performance, but it’s expensive, volatile, and has limited
scalability.
- Flash
storage (NAND) is non-volatile and less expensive than DRAM, but it lacks
DRAM’s performance.
- Spinning
hard-disk drives (HDDs) provide massive storage at the lowest price, but
physical discs bring well-understood total cost of ownership (TCO) issues
around reliability, physical space requirements, cooling, and power.
Collectively,
these traditional storage options have left significant gaps in the memory and
storage continuum in the data center, thus limiting the performance of
applications. Ever-increasing amounts of data, and the need to access more of
it quickly, have further magnified the problem.
In
particular, two memory and storage gaps stand out for organizations trying to
transform their data centers:
- Between
costly, low-capacity DRAM and more affordable NAND-based solid-state
drives (SSDs).
- Between
slower NAND SSDs and lower-cost, but less reliable, HDDs.
Organizations
have not had viable options that balance cost, capacity, and performance to
bridge those gaps—until now
UFS is a highly integrated solution which combines a feature-wise
flash controller compliant with the latest UFS2.1 standard and standard NAND
flash memory. Its high-performance storage accessing, better power efficiency,
and ease of system design make the UFS a fabulous solution for automotive,
industry, embedded and portable applications.
eMMC is optimally designed for a wide range of embedded
applications and is fully compliant to the JEDEC standards for eMMC*
4.5/5.0/5.1 protocols. Available in 100/153-ball BGA packages, eMMC eases PCB
design and enables low-cost manufacturing.
Features
·
Plug & Play only requires format/fdisk
prior to use
·
Small footprint for space-limited design
Lower total cost of ownership
·
Rugged & Reliable (no moving parts)
·
Eliminate requalification cost from NAND generation
change
·
Cost-saving with flexible TLCmode™/MLCmode™/SLCmode™, configurable capaities
Eliminate down time
·
Supports S.M.A.R.T. and advanced SSD Telemetry logging
features
·
IntelligentScan with DataRefresh for Data integrity enhancement
·
Full End-to-End data path protection with recovery
algorithms
·
Silicon Motion's 4th generation LDPC ECC engine with
Group Page RAID
·
Remote firmware update available
via secured digital signature
UFS/eMMC
incorporate full data error detection with recovery engines to provide enhanced
data integrity throughout the entire Host-to-NAND-to-Host data path. The
UFS/eMMC data recovery algorithm can effectively detect any error in the
UFS/eMMC data path, including hardware (i.e. ASIC)
errors, firmware errors, and memory errors arising in SRAM, DRAM, or NAND.
Conventional
UFS/eMMC employ standard BCH and RS ECC (error correction coding) engines to
initiate first-level correction using NAND shift-read-retries. In addition to
this first-level error correction, UFS/eMMC also implement a highly efficient
second-level correction scheme using an LDPC (low-density parity-check) code
and a Group page RAID algorithm (a highly efficient redundant backup) to reduce
potential DPPM at customer site while extending the service life of UFS/eMMC.
Our
proprietary Scan function will activate automatically to scan recharge, repair,
or retire the cell block (Data Refresh) according to the host behavior and
working environment (eg. ambient temperature). As a
result of the combination of Scan and Data Refresh, UFS/eMMC can
effectively prolong its service life much beyond typical NAND specifications.
UFS
is a highly integrated solution which combines a feature-wise flash controller
compliant with the latest UFS2.1 standard and standard NAND flash memory. Its
high-performance storage accessing, better power efficiency, and ease of system
design make the UFS a fabulous solution for automotive, industry, embedded and
portable applications. The UFS leverages industry leading technology and
experience in NAND management, and supports the UFS2.1 advanced features such as HS-Gear3 x 2-lane mode and command queue.
With extended temperature and various capacity support, offering easy and rapid
design integration, the UFS also ideally fits the requirements of point-of-sale
terminals, networking and telecommunications equipment, and a variety of
leading-edge industrial applications. With superior performance, multitasking
support, and high stability, the UFS can seamlessly serve the needs of a wide
variety of mobile devices and new booming embedded/portable applications.
Embedded applications using eMMC today can migrate to UFS for higher
performance and capacity options. Additionally, UFS can be customized via
firmware for specific features and applications. As the world’s leading NAND controller
vendor, Silicon Motion builds its products to the highest quality and
reliability standards – backed by outstanding sales and technical support from
design through post production. Silicon Motion’s commitment to automotive and
industrial quality is fully incorporated throughout the design, manufacturing
and qualification phases of its UFS products.
eMMC
is optimally designed for a wide range of embedded applications and is fully
compliant to the JEDEC standards for eMMC* 4.5/5.0/5.1 protocols. Available in
100/153-ball BGA packages, eMMC eases PCB design and enables low-cost
manufacturing. Built with industry proven controllers and high
quality NAND components, eMMC offers advanced NAND management features
including error correction, bad block management and health monitoring –
enabling the most highly reliable, non-volatile eMMC storage solution for
today’s cutting edge industrial, embedded and automotive applications. For
automotive IVI applications, eMMC features industry leading low DPPM, AEC-Q100
qualification, and product longevity support. Embedded applications using HDD
or raw NAND today can migrate to eMMC for higher performance and capacity
options. Additionally, eMMC can be customized via firmware for specific
features and applications. As the world’s leading NAND controller vendor,
Silicon Motion builds its products to the highest quality and reliability
standards – backed by uncompromised sales and technical support from design
through post production. Silicon Motion’s commitment to automotive and
industrial quality is fully incorporated throughout the design, manufacturing
and qualification phases of its eMMC products.
The
PCIe NVMe SSD IC family consists of first supporting
PCIe Gen3 x4 NVMe 1.3 and second supporting PCIe Gen3
x2 NVMe 1.3 designed optimally for high-performance
mission critical applications. By combining industry proven controller
technology, NAND flash and passive components into a small single BGA package,
The PCIe NVMe SSD IC simplifies design efforts,
reduces time-to-market while protecting from NAND technology migration
concerns. The SM689 supports embedded DRAM with Data Redundancy with PCIe Gen3
x4 interface - exhibiting sequential read speed of up to 1.6GB/s and sequential
write speed of up to 650MB/s. The SM681 DRAMLess
series feature the best balance of saving/performance - cost saving from
eliminating DRAM while maintaining DRAM-like performance via HMB (Host Memory
Buffer). Both available in 3D TLC/MLC/SLC modes, the unique flexible design can
support multiple capacity configurations ranging from 5GB to 480GB and include
enterprise-grade advanced data integrity and reliability capabilities using
Silicon Motion's proprietary end-to-end data protection, ECC and data caching
technologies. Key Features End to End Data Path Protection The
PCIe NVMe SSD IC incorporate full data error
detection with recovery engines to provide enhanced data integrity throughout
the entire Host-to-NAND-to-Host data path. The PCIe NVMe
SSD IC data recovery algorithm can effectively detect any error in the SSD data
path, including hardware (i.e. ASIC) errors, firmware
errors and memory errors arising in SRAM, DRAM or NAND.
NANDXtend™ ECC Engine
Conventional SSDs employ standard BCH and RS ECC (error correction coding)
engines for initiate first-level correction using NAND shift-read-retries. In
addition to this first-level error correction, The PCIe NVMe
SSD IC also implement a highly efficient second-level correction scheme using
an LDPC (low-density parity check) code and a Group page RAID algorithm (a
highly efficient redundant backup) to reduce potential DPPM at customer site
while extending the service life of SSD. No error data will be sent to host!
Write flow w / encode Host Encode CRC Write Engine Encode Decode ECC Engine
DRAM Encode Decode ECC Engine SRAM Encode Decode NAND ECC Engine RAID Engine
Read flow w / decode Host Decode CRC Engine Read Encode Decode ECC Engine DRAM
Encode Decode ECC Engine SRAM Encode Decode NAND ECC Engine RAID Engine The
PCIe NVMe SSD IC Higher ambient temp to increase Scan
frequency Thermo impact on NAND Data Retention 75.58 Mo 12 Mo 2.14 Mo Temp SLC
@ max PE MLC @ max PE 40 12 Mo 55 1.88 Mo 70 0.34 Mo 85 0.45 Mo 0.07 Mo Based
on Arrhenius Equation Scan/Data Refresh
to proactively extend Data Retention beyond the typical NAND flash limitation
Based on 1Ynm MLC, @ 1,000 PE Max ECC Correctable Preset / Programmable
Threshold 85℃ data retention simulation Data Refresh Error Bits Month 50 45 40
35 30 25 20 15 10 5 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57
60 63 66 69 72 75 78 81 84 87 90 93 96 99 102 105 108 111 114 117 120 Why The
PCIe NVMe SSD IC Specifications Easy to use ．Plug & Play only requires format/fdisk
prior to use ．Small footprint for space-limited design
Lower total cost of ownership ．Rugged & Reliable
(no moving parts) ．Eliminate requalification cost from
NAND generation change ．Cost saving with flexible
TLC/MLC/SLC modes, configurable capacities. Eliminate down time ．Support S.M.A.R.T. and advanced SSD Telemetry logging features ．
Scan with Data Refresh for Data integrity enhancement ．Full End-to-End data path protection with recovery algorithms ．SMI’s 4th generation LDPC ECC engine with
Group Page RAID ．Remote firmware update available via
secured digital signature Scan and Data Refresh to Enhance Data Integrity SMI’s
proprietary Scan
function will activate automatically to scan recharge, repair or retire the
cell block (Data Refresh) according to the host behavior and working
environment (eg. ambient temperature). As a result of
the combination of Scan and Data Refresh, The PCIe NVMe SSD IC can effective prolong its service life much
beyond typical NAND specifications.
Easy
to use ．
Plug
& Play only requires format/fdisk prior to use ．
Small
footprint for space-limited design
Lower
total cost of ownership ．
Rugged
& Reliable (no moving parts) ．
Eliminate
requalification cost from NAND generation change ．
Cost
saving with flexible TLC/MLC/SLC modes, configurable capacities.
Eliminate
down time ．
Support
S.M.A.R.T. and advanced SSD Telemetry logging features ．
Scan
with Data Refresh for Data integrity enhancement ．
Full
End-to-End data path protection with recovery algorithms ．
4th
generation LDPC ECC engine with Group Page RAID ．
Remote
firmware update available via secured digital
signature