U.S. patent application number 16/794404 was filed with the patent office on 2021-08-19 for information handling system having improved host memory buffer for input/output requests.
The applicant listed for this patent is DELL PRODUCTS, LP. Invention is credited to Young Hwan Jang, Lip Vui Kan.
Application Number | 20210255801 16/794404 |
Document ID | / |
Family ID | 1000005750143 |
Filed Date | 2021-08-19 |
United States Patent
Application |
20210255801 |
Kind Code |
A1 |
Jang; Young Hwan ; et
al. |
August 19, 2021 |
INFORMATION HANDLING SYSTEM HAVING IMPROVED HOST MEMORY BUFFER FOR
INPUT/OUTPUT REQUESTS
Abstract
A host memory buffer is configured as a write cache for
prioritizing read operations. When a peripheral device (such as a
solid-state drive) receives a mixture of read requests and write
requests, the write requests may be suspended or deferred to first
execute the read requests. The write requests may be cached to the
host memory buffer, thus allowing the peripheral device to process
the read requests for relatively immediate response. Once the read
requests are completed, the peripheral device may begin executing a
queue of the write requests stored to the host memory buffer. This
caching strategy results in a thirty percent (30%) improvement in
drive performance.
Inventors: |
Jang; Young Hwan;
(Gwangju-si, KR) ; Kan; Lip Vui; (Hillbrooks,
SG) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
DELL PRODUCTS, LP |
Round Rock |
TX |
US |
|
|
Family ID: |
1000005750143 |
Appl. No.: |
16/794404 |
Filed: |
February 19, 2020 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G06F 3/0611 20130101;
G06F 3/068 20130101; G06F 2212/1024 20130101; G06F 12/0866
20130101; G06F 3/0659 20130101 |
International
Class: |
G06F 3/06 20060101
G06F003/06; G06F 12/0866 20060101 G06F012/0866 |
Claims
1. A method, comprising: configuring, by a solid-state drive, an
external host memory buffer operating in a host memory device of a
host as a dedicated cache memory exclusively used by the
solid-state drive; receiving a read request sent via an
input/output channel from the host requesting the solid-state drive
to read a first data; receiving a write request sent via the
input/output channel from the host requesting the solid-state drive
to write a second data to an internal cache memory operating in the
solid-state drive; suspending an execution of the write request
received via the input/output channel from the host; prioritizing
the read request by caching the write request via the input/output
channel to the external host memory buffer operating in the host
memory device of the host as the dedicated cache memory exclusively
used by the solid-state drive; reading the first data by executing
the read request; in response to the executing of the read request,
retrieving the write request cached to the external host memory
buffer via the input/output channel from the host memory device of
the host as the dedicated cache memory exclusively used by the
solid-state drive; and executing the write request to write the
second data to the internal cache memory operating in the
solid-state drive.
2. The method of claim 1, further comprising retrieving a mapping
information associated with the read request.
3. The method of claim 1, further comprising retrieving a mapping
information associated with the write request.
4. The method of claim 1, further comprising executing an operating
system.
5. (canceled)
6. The method of claim 1, further comprising queuing the write
request.
7. The method of claim 1, further comprising sending the write
request to the external host memory buffer.
8. An information handling system comprising: a hardware processor;
and a memory device accessible to the hardware processor, the
memory device storing instructions that when executed by the
hardware processor perform operations, the operations including:
configuring a portion of the memory device as an external dedicated
cache memory exclusively used by a solid-state drive; sending a
read request via an input/output channel to the solid-state drive
requesting to read a first data from an internal cache memory
operating in the solid-state drive; sending a write request via the
input/output channel to the solid-state drive requesting to write a
second data to the internal cache memory operating in the
solid-state drive; receiving a cache request sent via the
input/output channel from the solid-state drive requesting to cache
the write request to the portion of the memory device configured as
the external dedicated cache memory exclusively used by the
solid-state drive; executing the read request to read the first
data; in response to the executing of the read request, retrieving
the write request cached to the portion of the memory device
configured as the external dedicated cache memory exclusively used
by the solid-state drive; and executing the write request to write
the second data to the internal cache memory operating in the
solid-state drive.
9. The system of claim 8, wherein the operations further include
retrieving a mapping information associated with the read
request.
10. The system of claim 8, wherein the operations further include
retrieving a mapping information associated with the write
request.
11. The system of claim 8, wherein the operations further include
executing an operating system.
12. The system of claim 8, wherein the operations further include
suspending the write request.
13. The system of claim 8, wherein the operations further include
queuing the write request.
14. (canceled)
15. A memory device storing instructions that when executed by a
processor perform operations, the operations including: configuring
a portion of a host memory buffer operating in a host as an
external dedicated cache memory exclusively used by a solid-state
drive; receiving a read request sent via an input/output channel
from the host requesting to read a first data from an internal
cache memory operating in the solid-state drive; receiving a write
request sent via the input/output channel from the host requesting
to write a second data to the internal cache memory operating in
the solid-state drive; suspending an execution of the write request
received via the input/output channel from the host; prioritizing
the read request by caching the write request via the input/output
channel to the portion of the host memory buffer operating in the
host as the external dedicated cache memory exclusively used by the
solid-state drive; in response to executing the read request,
retrieving the write request via the input/output channel from the
portion of the host memory buffer operating in the host as the
external dedicated cache memory exclusively used by the solid-state
drive; and executing the write request.
16. The memory device of claim 15, wherein the operations further
include retrieving a mapping information associated with the read
request.
17. The memory device of claim 15, wherein the operations further
include retrieving a mapping information associated with the write
request.
18. The memory device of claim 17, wherein the operations further
include executing an operating system.
19. (canceled)
20. The memory device of claim 15, wherein the operations further
include queuing the write request.
Description
FIELD OF THE DISCLOSURE
[0001] This disclosure generally relates to information handling
systems, and more particularly relates to redirecting write
operations to host memory buffers and/or to prioritizing read
operations in read/write mixed input/output requests.
BACKGROUND
[0002] As the value and use of information continues to increase,
individuals and businesses seek additional ways to process and
store information. One option is an information handling system. An
information handling system generally processes, compiles, stores,
and/or communicates information or data for business, personal, or
other purposes. Because technology and information handling needs
and requirements may vary between different applications,
information handling systems may also vary regarding what
information is handled, how the information is handled, how much
information is processed, stored, or communicated, and how quickly
and efficiently the information may be processed, stored, or
communicated. The variations in information handling systems allow
for information handling systems to be general or configured for a
specific user or specific use such as financial transaction
processing, reservations, enterprise data storage, or global
communications. In addition, information handling systems may
include a variety of hardware and software resources that may be
configured to process, store, and communicate information and may
include one or more computer systems, data storage systems, and
networking systems.
SUMMARY
[0003] A host memory buffer is configured as a write cache for
prioritizing read operations. When a peripheral device (such as a
solid-state drive) receives a mixture of read requests and write
requests, the write requests may be suspended or deferred to first
execute the read requests. The write requests may thus be cached to
the host memory buffer, thus allowing the peripheral device to
process the read requests for relatively immediate response. Once
the read requests are completed, the peripheral device may begin
executing a queue of the write requests stored to the host memory
buffer. This caching strategy results in a thirty percent (30%)
improvement in drive performance.
BRIEF DESCRIPTION OF THE DRAWINGS
[0004] It will be appreciated that for simplicity and clarity of
illustration, elements illustrated in the Figures have not
necessarily been drawn to scale. For example, the dimensions of
some of the elements are exaggerated relative to other elements.
Embodiments incorporating teachings of the present disclosure are
shown and described with respect to the drawings presented herein,
in which:
[0005] FIG. 1 is a block diagram of a generalized information
handling system;
[0006] FIGS. 2-3 illustrate a host memory buffer, according to
exemplary embodiments;
[0007] FIGS. 4-7 illustrate a caching method, according to
exemplary embodiments;
[0008] FIGS. 8-9 illustrate performance improvements, according to
exemplary embodiments; and
[0009] FIG. 10 illustrates a method or algorithm for caching to the
host memory buffer, according to exemplary embodiments.
[0010] The use of the same reference symbols in different drawings
indicates similar or identical items.
DETAILED DESCRIPTION OF DRAWINGS
[0011] The following description in combination with the Figures is
provided to assist in understanding the teachings disclosed herein.
The following discussion will focus on specific implementations and
embodiments of the teachings. This focus is provided to assist in
describing the teachings, and should not be interpreted as a
limitation on the scope or applicability of the teachings.
[0012] FIG. 1 illustrates an embodiment of an information handling
system 100 including processors 102 and 104, chipset 110, memory
120, graphics adapter 130 connected to video display 134,
non-volatile RAM (NV-RAM) 140 that includes a basic input and
output system/extensible firmware interface (BIOS/EFI) module 142,
disk controller 150, hard disk drive (HDD) 154, optical disk drive
(ODD) 156, disk emulator 160 connected to solid state drive (SSD)
164, an input/output (I/O) interface 170 connected to an add-on
resource 174, a trusted platform module (TPM) 176, a network
interface device 180, and a baseboard management controller (BMC)
190. Processor 102 is connected to chipset 110 via processor
interface 106, and processor 104 is connected to chipset 110 via
processor interface 108.
[0013] Chipset 110 represents an integrated circuit or group of
integrated circuits that manages data flow between processors 102
and 104 and the other elements of information handling system 100.
In a particular embodiment, chipset 110 represents a pair of
integrated circuits, such as a north bridge component and a south
bridge component. In another embodiment, some or all of the
functions and features of chipset 110 are integrated with one or
more of processors 102 and 104. Memory 120 is connected to chipset
110 via a memory interface 122. An example of memory interface 122
includes a Double Data Rate (DDR) memory channel, and memory 120
represents one or more DDR Dual In-Line Memory Modules (DIMMs). In
a particular embodiment, memory interface 122 represents two or
more DDR channels. In another embodiment, one or more of processors
102 and 104 include memory interface 122 that provides a dedicated
memory for the processors. A DDR channel and the connected DDR
DIMMs can be in accordance with a particular DDR standard, such as
a DDR3 standard, a DDR4 standard, a DDR5 standard, or the like.
Memory 120 may further represent various combinations of memory
types, such as Dynamic Random Access Memory (DRAM) DIMMs, Static
Random Access Memory (SRAM) DIMMs, non-volatile DIMMs (NV-DIMMs),
storage class memory devices, Read-Only Memory (ROM) devices, or
the like.
[0014] Graphics adapter 130 is connected to chipset 110 via a
graphics interface 132, and provides a video display output 136 to
a video display 134. An example of a graphics interface 132
includes a peripheral component interconnect-express interface
(PCIe) and graphics adapter 130 can include a four lane (.times.4)
PCIe adapter, an eight lane (.times.8) PCIe adapter, a 16-lane
(.times.16) PCIe adapter, or another configuration, as needed or
desired. In a particular embodiment, graphics adapter 130 is
provided on a system printed circuit board (PCB). Video display
output 136 can include a digital video interface (DVI), a high
definition multimedia interface (HDMI), DisplayPort interface, or
the like. Video display 134 can include a monitor, a smart
television, an embedded display such as a laptop computer display,
or the like.
[0015] NV-RAM 140, disk controller 150, and I/O interface 170 are
connected to chipset 110 via I/O channel 112. An example of I/O
channel 112 includes one or more point-to-point PCIe links between
chipset 110 and each of NV-RAM 140, disk controller 150, and I/O
interface 170. Chipset 110 can also include one or more other I/O
interfaces, including an Industry Standard Architecture (ISA)
interface, a Small Computer Serial Interface (SCSI) interface, an
Inter-Integrated Circuit (I.sup.2C) interface, a System Packet
Interface (SPI), a Universal Serial Bus (USB), another interface,
or a combination thereof. NV-RAM 140 includes BIOS/EFI module 142
that stores machine-executable code (BIOS/EFI code) that operates
to detect the resources of information handling system 100, to
provide drivers for the resources, to initialize the resources, and
to provide common access mechanisms for the resources. The
functions and features of BIOS/EFI module 142 will be further
described below.
[0016] Disk controller 150 includes a disk interface 152 that
connects the disc controller 150 to HDD 154, to ODD 156, and to
disk emulator 160. Disk interface 152 may include an integrated
drive electronics (IDE) interface, an advanced technology
attachment (ATA) such as a parallel ATA (PATA) interface or a
serial ATA (SATA) interface, a SCSI interface, a USB interface, a
proprietary interface, or a combination thereof. Disk emulator 160
permits a solid-state drive (SSD) 164 to be connected to
information handling system 100 via an external interface 162. An
example of external interface 162 includes a USB interface, an IEEE
1394 (Firewire) interface, a proprietary interface, or a
combination thereof. Alternatively, SSD 164 can be disposed within
information handling system 100.
[0017] I/O interface 170 includes a peripheral interface 172 that
connects I/O interface 170 to add-on resource 174, to TPM 176, and
to network interface device 180. Peripheral interface 172 can be
the same type of interface as I/O channel 112, or can be a
different type of interface. As such, I/O interface 170 extends the
capacity of I/O channel 112 when peripheral interface 172 and the
I/O channel are of the same type, and the I/O interface translates
information from a format suitable to the I/O channel to a format
suitable to the peripheral channel 172 when they are of a different
type. Add-on resource 174 can include a sound card, data storage
system, an additional graphics interface, another add-on resource,
or a combination thereof. Add-on resource 174 can be on a main
circuit board, a separate circuit board or an add-in card disposed
within information handling system 100, a device that is external
to the information handling system, or a combination thereof.
[0018] Network interface device 180 represents a network
communication device disposed within information handling system
100, on a main circuit board of the information handling system,
integrated onto another element such as chipset 110, in another
suitable location, or a combination thereof. Network interface
device 180 includes a network channel 182 that provides an
interface to devices that are external to information handling
system 100. In a particular embodiment, network channel 182 is of a
different type than peripheral channel 172 and network interface
device 180 translates information from a format suitable to the
peripheral channel to a format suitable to external devices. In a
particular embodiment, network interface device 180 includes a host
bus adapter (HBA), a host channel adapter, a network interface card
(NIC), or other hardware circuit that can connect the information
handling system to a network. An example of network channel 182
includes an InfiniBand channel, a fiber channel, a gigabit Ethernet
channel, a proprietary channel architecture, or a combination
thereof. Network channel 182 can be connected to an external
network resource (not illustrated). The network resource can
include another information handling system, a data storage system,
another network, a grid management system, another suitable
resource, or a combination thereof.
[0019] BMC 190 is connected to multiple elements of information
handling system 100 via one or more management interface 192 to
provide out of band monitoring, maintenance, and control of the
elements of the information handling system. As such, BMC 190
represents a processing device different from processors 102 and
104, which provides various management functions for information
handling system 100. In an embodiment, BMC 190 may be responsible
for granting access to a remote management system that may
establish control of the elements to implement power management,
cooling management, storage management, and the like. The BMC 190
may also grant access to an external device. In this case, the BMC
may include transceiver circuitry to establish wireless
communications with the external device such as a mobile device.
The transceiver circuitry may operate on a Wi-Fi channel, a
near-field communication (NFC) channel, a Bluetooth or
Bluetooth-Low-Energy (BLE) channel, a cellular based interface such
as a global system for mobile (GSM) interface, a code-division
multiple access (CDMA) interface, a universal mobile
telecommunications system (UMTS) interface, a long-term evolution
(LTE) interface, another cellular based interface, or a combination
thereof. A mobile device may include Ultrabook, a tablet computer,
a netbook, a notebook computer, a laptop computer, mobile
telephone, a cellular telephone, a smartphone, a personal digital
assistant, a multimedia playback device, a digital music player, a
digital video player, a navigational device, a digital camera, and
the like.
[0020] The term BMC may be used in the context of server systems,
while in a consumer-level device a BMC may be referred to as an
embedded controller (EC). A BMC included at a data storage system
can be referred to as a storage enclosure processor. A BMC included
at a chassis of a blade server can be referred to as a chassis
management controller, and embedded controllers included at the
blades of the blade server can be referred to as blade management
controllers. Out-of-band communication interfaces between BMC and
elements of the information handling system may be provided by
management interface 192 that may include an inter-integrated
circuit (I2C) bus, a system management bus (SMBUS), a power
management bus (PMBUS), a low pin count (LPC) interface, a serial
bus such as a universal serial bus (USB) or a serial peripheral
interface (SPI), a network interface such as an Ethernet interface,
a high-speed serial data link such as PCIe interface, a network
controller-sideband interface (NC-SI), or the like. As used herein,
out-of-band access refers to operations performed apart from a
BIOS/operating system execution environment on information handling
system 100, that is apart from the execution of code by processors
102 and 104 and procedures that are implemented on the information
handling system in response to the executed code.
[0021] In an embodiment, the BMC 190 implements an integrated
remote access controller (iDRAC) that operates to monitor and
maintain system firmware, such as code stored in BIOS/EFI module
142, option ROMs for graphics interface 130, disk controller 150,
add-on resource 174, network interface 180, or other elements of
information handling system 100, as needed or desired. In
particular, BMC 190 includes a network interface 194 that can be
connected to a remote management system to receive firmware
updates, as needed or desired. Here BMC 190 receives the firmware
updates, stores the updates to a data storage device associated
with the BMC, transfers the firmware updates to NV-RAM of the
device or system that is the subject of the firmware update,
thereby replacing the currently operating firmware associated with
the device or system, and reboots information handling system,
whereupon the device or system utilizes the updated firmware
image.
[0022] BMC 190 utilizes various protocols and application
programming interfaces (APIs) to direct and control the processes
for monitoring and maintaining the system firmware. An example of a
protocol or API for monitoring and maintaining the system firmware
includes a graphical user interface (GUI) associated with BMC 190,
an interface defined by the Distributed Management Taskforce (DMTF)
(such as Web Services Management (WS-MAN) interface, a Management
Component Transport Protocol (MCTP) or, Redfish interface), various
vendor defined interfaces (such as Dell EMC Remote Access
Controller Administrator (RACADM) utility, Dell EMC Open Manage
Server Administrator (OMSS) utility, Dell EMC Open Manage Storage
Services (OMSS) utility, Dell EMC Open Manage Deployment Toolkit
(DTK) suite), representational state transfer (REST) web API, a
BIOS setup utility such as invoked by a "F2" boot option, or
another protocol or API, as needed or desired.
[0023] In a particular embodiment, BMC 190 is included on a main
circuit board (such as a baseboard, a motherboard, or any
combination thereof) of information handling system 100, or is
integrated into another element of the information handling system
such as chipset 110, or another suitable element, as needed or
desired. As such, BMC 190 can be part of an integrated circuit or a
chip set within information handling system 100. BMC 190 may
operate on a separate power plane from other resources in
information handling system 100. Thus BMC 190 can communicate with
the remote management system via network interface 194 or the BMC
can communicate with the external mobile device using its own
transceiver circuitry while the resources or elements of
information handling system 100 are powered off or at least in low
power mode. Here, information can be sent from the remote
management system or external mobile device to BMC 190 and the
information can be stored in a RAM or NV-RAM associated with the
BMC. Information stored in the RAM may be lost after power-down of
the power plane for BMC 190, while information stored in the NV-RAM
may be saved through a power-down/power-up cycle of the power plane
for the BMC.
[0024] In a typical usage case, information handling system 100
represents an enterprise class processing system, such as may be
found in a datacenter or other compute-intense processing
environment. Here, there may be hundreds or thousands of other
enterprise class processing systems in the datacenter. In such an
environment, the information handling system may represent one of a
wide variety of different types of equipment that perform the main
processing tasks of the datacenter, such as modular blade servers,
switching and routing equipment (network routers, top-of-rack
switches, and the like), data storage equipment (storage servers,
network attached storage, storage area networks, and the like), or
other computing equipment that the datacenter uses to perform the
processing tasks.
[0025] For purposes of this disclosure information handling system
100 can include any instrumentality or aggregate of
instrumentalities operable to compute, classify, process, transmit,
receive, retrieve, originate, switch, store, display, manifest,
detect, record, reproduce, handle, or utilize any form of
information, intelligence, or data for business, scientific,
control, entertainment, or other purposes. For example, information
handling system 100 can be a personal computer, a laptop computer,
a smart phone, a tablet device or other consumer electronic device,
a network server, a network storage device, a switch router or
other network communication device, or any other suitable device
and may vary in size, shape, performance, functionality, and price.
Further, information handling system 100 can include processing
resources for executing machine-executable code, such as a central
processing unit (CPU), a programmable logic array (PLA), an
embedded device such as a System-on-a-Chip (SoC), or other control
logic hardware. Information handling system 100 can also include
one or more computer-readable medium for storing machine-executable
code, such as software or data. Additional components of
information handling system 100 can include one or more storage
devices that can store machine-executable code, one or more
communications ports for communicating with external devices, and
various input and output (I/O) devices, such as a keyboard, a
mouse, and a video display. Information handling system 100 can
also include one or more buses operable to transmit information
between the various hardware components.
[0026] FIGS. 2-3 illustrate a host memory buffer 200, according to
exemplary embodiments. The host memory buffer 200 may be any
portion of the memory 120 that is configured for, and/or dedicated
to, the solid-state drive (SSD) 164. When the processors 102/104
access and/or execute an operating system 202 (perhaps stored by
the memory 120), the operating system 202 may allocate and/or
partition the host memory buffer 200 to the solid-state drive 164.
The solid-state drive 164 may have exclusive use of the host memory
buffer 200, or the host memory buffer 200 may be a shared resource
with other components, peripherals, or software/services. Even
though the solid-state drive 164 may have its own internal
processor and memory device (not shown for simplicity), the host
memory buffer 200 may increase speed and efficiency for some
input/output tasks. The host memory buffer 200 takes advantage of
the I/O channel 112 (such as the direct memory access capabilities
of PCI Express) to allow the solid-state drive 164 to access and
use any of the memory 120 (such as DRAM).
[0027] FIG. 3 illustrates the host memory buffer 200 as a cache
memory 204. Conventional schemes utilize the host memory buffer 200
to cache mapping information (such as a mapping table of logical
block addresses to physical NAND addresses). This disclosure,
though, uses the host memory buffer 200 as write cache for read
prioritization. Because the size of the host memory buffer 200 is
substantially greater (i.e., 60 times larger) than the conventional
cache internal to the solid-state drive 164, the host memory buffer
200 may efficiently be used to prioritize read requests or
operations 206 over write requests or operations 208 in a typical
client busty workload patterns. When the host (such as the
information handling system 100) issues read and write requests 206
and 208 in a mixed sequence, the solid-state drive 164 may
re-schedule any write operation 208 from its internal cache to the
host memory buffer 200. The solid-state drive 164 may thus
prioritize the read operation 206 without first completing and/or
executing the write operation 208. This caching strategy extracts
maximum potential read write performance in typical mixed workload
conditions.
[0028] FIGS. 4-7 illustrate the caching strategy, according to
exemplary embodiments. FIGS. 4-5 illustrate read/write bottlenecks.
When the information handling system 100 issues or sends a mixed
sequence of the read operations 206 and the write operations 208, a
new or recent read request or operation 206 is blocked, or delayed,
by a prior or historical write request or operation 208 that is
currently being executed. The read request 206, in other words, may
be substantially delayed until the write request 208 is completed.
The mixed sequence (of the read operations 206 and the write
operations 208) thus causes a software program to be delayed. Data
writes, in other words, may require longer times than data reads.
When the software program is in process at the component media
(such as the solid-state drive 164), any read operation 206 has to
be waited until NAND media completes the write operation 208. FIG.
5 illustrates this read/write performance limitation by charting
NVMe SSD performance in different ratios of read and write I/O,
swinging from 100% read to 100% write in mixed manner (tested with
Hynix PC400 512 GB, 4 KB random read/write mixed at Queue Depth 4,
at 8 GB range). When the read and write mixed ratio is 75% and 25%,
the combined performance is about thirty percent (30%) less than
ideally perform based on its raw read (100%) and write numbers.
Real life workloads are always mixture of random read/write
operations. So, the real-life benchmark numbers such as PCMark are
often less than those synthetic benchmark numbers for today's
SSDs.
[0029] FIGS. 6-7, though, illustrate prioritization. When the
solid-state drive 164 receives any write request 208, FIG. 6
illustrates the solid-state drive 164 suspending execution of the
write request 208. The solid-state drive 164, instead, may send
and/or cache the write request 208 to the host memory buffer 200.
Because the controller of the solid-state drive 164 need not
immediately execute the write request 208, the solid-state drive
164 may ignore or defer the write request 208 and, instead,
prioritize the read request 206 over the write request 208. The
solid-state drive 164 may thus consult any mapping table or
information, retrieve the corresponding data blocks to the read,
and return send the data blocks to the host information handling
system 100 in response to the read request 206. As FIG. 7
illustrates, when the solid-state drive 164 is idle (perhaps when
all pending read requests 206 have been completed), the controller
of the solid-state drive 164 may begin executing a queue of the
write requests 208. The solid-state drive 164 may thus begin
retrieving and executing the queued write requests 208 that were
previously cached to the host memory buffer 200. The solid-state
drive 164 thus flushes the pending write requests 208 in any order
(such as chronological FIFO or LIFO). As each write request 208 is
completed, the solid-state drive 164 may update any mapping table
or information with logical and physical addresses. Should the
solid-state drive 164 receive another or intervening read request
206, the solid state drive 164 may again suspend or defer the
remaining cache of the write requests 208 stored in the host memory
buffer 200.
[0030] FIGS. 8-9 illustrate performance improvements, according to
exemplary embodiments. Because the read operation 206 is
prioritized, machine performance and the user's experience are
enhanced. The host memory buffer 200 optimizes the input/output
schedule in real life mixed IO scenarios. Any trace-based benchmark
will result up to 30% better performance using the same SSD
hardware. There is no additional cost since it is device firmware
implementation. FIGS. 8-9 thus illustrate PC Mark 8 benchmarking
(trace-based) analysis and simulation results. Real life workload
consists of small payload sizes, with 4 KB is the highest used
payload size. Access patterns are mixture of read and write (80%
reads and 20% writes). Most of queue depths are low (less than 4),
and 95% of the queue depths are less than eight (8). As FIG. 9
shows, the solid-state drive 164 with the host memory buffer 200
for write cache plus read prioritization boosts the performance of
the same drive by 5%. The early simulated performance result is
only able to prioritized 10% of potential read in the mixed
workload. With further optimization, it can help to boost the
performance up to 30%. This means that the write caching strategy
differentiates system performance by 30% at the same hardware
configuration.
[0031] Exemplary embodiments thus present an elegant solution. The
solid-state drive 164 may suspend execution of any write request
208 and, instead, send or cache the write request 208 to the host
memory buffer 200. The solid-state drive 164 may thus prioritize
the read request 206 for relatively immediate response. Should no
read requests 206 be pending, the solid-state drive 164 may then
change its caching strategy and begin executing the queue of the
write requests 208 stored in the host memory buffer 200. A thirty
percent (30%) improvement in drive performance may result.
[0032] FIG. 10 illustrates a method or algorithm for caching to the
host memory buffer 200, according to exemplary embodiments. The
host information handling system 100 sends a sequence of
input/output requests to the solid-state drive 164 (Block 220). The
byte sizes of any of input/output requests (such as the read
operation 206 and/or the write operation 208) are compared to a
threshold size (such as 16 KB). If the payload sizes of the
input/output requests fail to satisfy the threshold size (Block
222), then the input/output requests (such as read commands 206
and/or the write commands 208) are queued/scheduled (Block 224) and
executed (Block 226). While the threshold size may be configurable
to suit performance and memory objectives, large payloads (such as
greater than the 16 KB) may be unsuited for smaller sizes of the
host memory buffer 200 (such as 64 MB). However, should the payload
sizes of the input/output requests satisfy the threshold size
(Block 222), then the amount of the data to be written (as
described or specified by the write request 208) is compared to the
byte size of the host memory buffer 200. If the amount of the data
to be written is less than size of the host memory buffer 200
(Block 228), then the read requests 206 may be split from the write
requests 208 and sequentially queued (Block 230). The write
requests 208 are written and queued to the host memory buffer 200
(Block 232). However, if the amount of the data to be written is
greater than size of the host memory buffer 200 (Block 228), then
the write requests 208 are written and queued to the host memory
buffer 200 (Block 232). When the read requests 206 are completed,
the queued write requests 208 (in the host memory buffer 200) are
flushed and written to the NAND of the solid-state drive 164 (Block
234).
[0033] Although only a few exemplary embodiments have been
described in detail herein, those skilled in the art will readily
appreciate that many modifications are possible in the exemplary
embodiments without materially departing from the novel teachings
and advantages of the embodiments of the present disclosure.
Accordingly, all such modifications are intended to be included
within the scope of the embodiments of the present disclosure as
defined in the following claims. In the claims, means-plus-function
clauses are intended to cover the structures described herein as
performing the recited function and not only structural
equivalents.
[0034] Devices, modules, resources, or programs that are in
communication with one another need not be in continuous
communication with each other, unless expressly specified
otherwise. In addition, devices, modules, resources, or programs
that are in communication with one another can communicate directly
or indirectly through one or more intermediaries.
[0035] The above-disclosed subject matter is to be considered
illustrative, and not restrictive, and the appended claims are
intended to cover any and all such modifications, enhancements, and
other embodiments that fall within the scope of the present
invention. Thus, to the maximum extent allowed by law, the scope of
the present invention is to be determined by the broadest
permissible interpretation of the following claims and their
equivalents, and shall not be restricted or limited by the
foregoing detailed description.
* * * * *