Finding the best SSD (Solid State Drive) Laptops
An SSD is a powerful pacemaker for any notebook. Techtestreport presents the most interesting laptop models with such a data turbo. In the video we present a great explanation of the different types of storage memories.
What SSDs can do
SSDs (Solid State Drive) have three characteristics, which makes them better compared to hard disks – and these should be considered when buying a notebook:
Speed: As flash memory, SSDs deliver data much faster than an HDD. Fast conventional hard disks often remain below 100 MByte/s when reading data, for example, whereas SSD values are typically around 500 MByte/s. When writing, older models sometimes achieve less than 200 MByte/s, but much higher values of around 500 MByte/s are also typical for the current best-performing SSDs.
Low power consumption: The low power consumption of SSDs also qualifies them as ideal mass storage devices in notebooks. Most SSDs consume less than 1.0 watts in operation. The minimum for HDDs is three to four times this amount.
Silence: Since there are no moving parts, an SSD works silently – in contrast to standard storage, which is one of the main noise generators in a notebook, along with a fan that runs out of true.
We tested three top Laptops equipped with very good SSD memory drives at different price points. You don’t have to spend a fortune to get a good laptop with a fast SSD onboard.
Test Results: Best SSD (Solid State Drive) Laptops
Ranking First: Dell XPS 15
- Doubles as a capable gaming PC
- 1 TerraByte SSD
- Good value despite high pricing
- 1080p and 4K display options
- Up to 32GB RAM
- Not the cheapest option
Best performing Gaming Laptop with a fast SSD
Normally, in comparative tests, product A from manufacturer X competes with product B from manufacturer Y. Especially in the case of notebooks, the fact that practically every model is available in various configurations is somewhat neglected – and often the top version costs twice as much as the entry-level version.
However, only one device with a specific configuration is tested – and connect is no exception. Against this background, we have carried out a comparative test of a different kind: The 15 inch convertible from Dell’s popular XPS family practically competes against itself.
On the one hand the standard configuration with a Full HD display, Core i5 processor, 8 GB RAM and a 256 GB SSD, on the other hand the top model with a high-resolution 4K screen, Core i7 CPU and twice as much RAM as data memory.
Both variants belong to the mobile upper class in monetary terms, but 700 Dollar lie in between. Which automatically raises the question to what extent the price difference is justified.
As so often, there is no clear answer to this question. But one by one. Let’s start with the display, the most significant difference between the two convertibles (convertible because the display can be turned 360 degrees).
4K corresponds to a resolution of 3840 x 2160 pixels – compared to FullHD, this means twice as much pixel density, in total four times as many pixels are illuminated. Of course, this does not remain without consequences. In order not to give the wrong impression: There’s nothing wrong with the FullHD display.
But if you put both devices next to each other, the difference becomes immediately visible. On the 4K screen, the content appears much sharper and even a good bit brighter, in addition to an impressive color brilliance. This point clearly goes to the top variant.
The other differences in equipment weigh less heavily. The larger RAM and the better processor do not make themselves felt too much in terms of performance – in these cases it is enough to get the maximum score in this category.
In many areas the close relationship leads to identical ratings without any difference.
This applies to the lavish interface supply (four fast USB-C ports, two of them with Thunderbolt-3 support, plus a USB-A adapter), the excellent workmanship, the high-quality input devices, the fingerprint scanner for Windows Hello hidden in the power button, and the sensational graphics performance, which is unparalleled in the ultrabook segment.
An unexpected difference between the siblings to this extent resulted from the endurance test. This is because the cheaper variant lasts about five and a half hours longer than the more expensive one in our runtime simulation, although the same 75 Wh battery is built into both devices.
The reason for this, apart from the slightly higher power consumption of the higher-quality computing unit, is above all the significantly higher energy hunger of the 4K display compared to its Full HD counterpart.
Verdict: Best performing Gaming Laptop with a fast SSD
The fact that the cheaper variant is slightly ahead in our test at the end was not necessarily to be expected, but it is conclusive. The high-end model has clear advantages in terms of features, but they don’t have an overly strong impact in terms of points, as for example the superiority of the display is hardly quantifiable.
It looks different with the large difference in battery life, which is fully reflected in the result – not least because the mobility factor is given special weight in our Ultrabook test procedure.
In view of the price advantage, the editors’ purchase recommendation goes to the Full-HD variant. However, those who place special value on an extremely bright, razor-sharp picture are advised to buy the 4K version despite the significant surcharge. Both are very good mobile computers with great SSDs onboard.
Ranking Second: Asus Zenbook 14
- Compact and durable design
- Solid productivity performance
- Luxurious aluminum chassis
- Best price-performance ratio
- 512 GB SSD
- No Thunderbolt 3
Best price-performance ratio Laptop with a fast SSD
The screenpad 2.0 is no longer reserved for the Zenbook Pro. Asus also sells the Zenbook 14 and Zenbook 15 with the second screen in the touchpad. Moreover, the case has been reworked and made considerably more compact. Nevertheless, a dedicated GPU fits in.
Asus will present its improved Zenbooks before the Ifa 2019: The Zenbook 14 and the Zenbook 15 mainly differ in their display sizes of 14 and 15.6 inches, respectively. Accordingly, the chassis of the 14 inch model measures 12.60 x 7.80 x 0.70 inches. Compared to the two-year-old predecessor from 2017, the models have about 13 percent less floor space, which is partly due to the somewhat narrower display edges. The new versions also no longer have an unused edge below the screen.
The screenpad 2.0 serves as a second screen or for displaying various Asus applications, such as a calendar, a calculator or a simple number pad. The 5.65-inch toucscreen proved to be helpful in the Zenbook Pro 14 test, but it requires a lot of energy.
Both notebooks use a reflective IPS panel with Full HD resolution. There are optional versions with a touchscreen, and the 15-inch model also has a 4K panel. A dedicated Geforce GTX 1650 GPU only fits into the larger version. 14-inch variant uses a Geforce MX 250, which computes more slowly. Both notebooks are available with Core i5-8265U or Core i7-8565U, each with four cores, eight threads and 1.8 or 1.6 GHz clock rate. The 15 Watt Whiskey Lake chips are already a bit older. In addition, there is eight or 16 GByte DDR4 memory and an M.2 SSD with 256, 512 or 1,000 GByte memory.
Both notebooks have enough connections
Both notebooks have a similar connection variety: Two USB-A sockets and one USB-C port with 3.2 Gen 2 speed as well as one HDMI port. The 14-inch device has a micro SD card reader, the 15.6-inch notebook an SD card reader. Both systems use Wi-Fi 5 and Bluetooth 5.0. The battery has a capacity of 50 and 72 watt hours respectively.
Verdict: Best price-performance ratio Laptop with a fast SSD
Asus already sells both notebooks. The Zenbook 14 costs from 1.100 Dollar, the Zenbook 15 is sold for 1.700 Dollar. Both are great laptops with a fast & high-capacity SSD onboard. The pricing is very fair which makes the ZenBook 14 our price-performance ratio winner for Laptops with a fast SSD.
Ranking Third: Acer Swift 5
- Ultra-thin and lightweight bod
- Very fast and responsive device
- Good IPS display with touchscreen
- 256 GB SSD
- Lack of RJ45 port
Great performing Laptop with a fast SSD at a cheaper price
The current Acer Swift 5 weighs only 2.2 lbs. despite its dimensions of 14.09 x 9.06 x 0.63 inches. However, the case is made entirely of metal – the rare magnesium-lithium alloy makes it almost as light as a feather. Acer has done a lot right in terms of workmanship: The high-quality case is complemented by an excellent keyboard and a good touchpad. The keyboard offers a pleasant, short stroke distance and a good key resistance. Therefore, it is also very suitable for frequent typists. The touchpad reacts precisely to movements, so that the pointer moves smoothly in circles across the screen. The keystroke via touchpad also works reliably – but the clicking noise is quite loud.
Unfortunately, Acer has refrained from bevelling the case edge on the front, because then it would be ideal as a wrist-rest. Instead, it can happen that the edge presses unpleasantly into the wrist during longer work on the notebook.
Acer Swift 5 with excellent performance rates
Those who choose this lightweight do not have to sacrifice good performance values. This is because an Intel Core i7-8565U from the current Whiskey Lake generation works inside the Acer Swift 5. This four-core works at 1.8 GHz in standard mode and even reaches a full 4.6 GHz thanks to Turbo Boost. An integrated Intel UHD Graphics 620 is used as the graphics solution. This not only provides enough power for MS Office applications, but you can also watch videos in Full-HD. However, the graphics card doesn’t have its own video memory, but always uses a part of the 8 GByte RAM. Here Acer relies on a DDR4 2400 RAM module. Because the notebook only has one slot, you can’t add any more memory bars.
In order to provide you with enough memory for Windows, your applications and your personal data, Acer relies on a 256 GByte SSD. To transfer data at a good speed, the SSD uses the PCI Express port.
The Acer Swift 5 offers many connection options. Besides a jack socket for a headset, you’ll also find three USB 3.x ports on the notebook – two of them are USB 3.0 type and one USB 3.1 type-C with a display port feature. In addition, there is also a classic HDMI connection. So that you can be on the road quickly, the notebook works with the WLAN-802.11ac standard. However, you have to do without an Ethernet connection. If you want to connect speakers, smartphones or tablets to the notebook, you can do this via Bluetooth 4.0.
Perfect for on the road, not only thanks to its low weight
Despite its light weight, the Acer Swift 5 has a 15.6-inch display diagonal and, thanks to the IPS Pro display, has a resolution of 1920 x 1080 pixels – in other words, offers Full HD resolution. In addition, the screen has good touch functionality, so you can also operate the notebook in this way. However, the display is also the device’s Achilles’ heel, as the maximum brightness of 299.6 candelas per square meter is rather among the weaker ones in the CHIP best list of premium notebooks. The checkerboard contrast of 179:1 is okay, but doesn’t reach any best values. In return, the display scores with 98.1% of the sRGB color space coverage and still 74.4% in the AdobeRGB color space measurement.
With its 2.2 lbs., the Acer Swift 5 is a lightweight and is therefore perfectly suited as a notebook that you can carry around with you in all situations. The manufacturer even dispenses with chunky components in the power supply unit and only supplies a slightly larger power plug. Moreover, the battery life can also score points when playing videos as well as when working with various office applications. If you only want to use the notebook for playing videos, then one battery charge is sufficient for about 9 hours and 30 minutes, while the runtime in office mode is 5 hours and 40 minutes. Neither of these are peak values, but especially the video runtime should be sufficient for most applications.
Verdict: Great performing Laptop with a fast SSD at a cheaper price
The Acer Swift 5 could convince us in the test. Especially the low weight of around 2.2 lbs. in combination with the good to very good test scores in every category are powerful arguments. Compared to the strong performance, we also find the price very fair. However, the somewhat weaker measured values in the display, which only offers Full-HD resolution in comparison to other premium notebooks, as well as some missing connections prevent a top placement in our best list. All in all the Acer Swift 5 offers great perfromance with a very capable 256GB SSD onborad.
Solid-State Drive (SSD) – What it is & How it works
An SSD (Solid State Drive) is a type of non-volatile storage medium that permanently stores data on a solid-state flash memory. Two key components form an SSD: a flash controller and NAND flash memory chips. The architectural configuration of the SSD controller is optimized to achieve high read and write performance for both sequential and random data requests. SSDs are sometimes referred to as flash drives or solid state disks.
Unlike a hard disk drive (HDD), an SSD has no moving parts that can break or move up or down. A conventional HDD consists of a rotating disk with a read/write head on a mechanical arm called an actuator. The HDD mechanism and the HDD are packaged as an integrated unit. In the past, companies and computer manufacturers have used rotating disks because of their lower unit cost and higher average lifespan, although SSDs are now common in desktop and laptop PCs.
How SSDs work
A rotating hard disk reads and writes data magnetically and is one of the oldest storage media in continuous operation. However, its magnetic properties can lead to mechanical failures. An SSD, on the other hand, reads and writes data to a substrate of interconnected flash memory chips made of silicon. Manufacturers build SSDs by stacking the chips in a grid to achieve different densities.
To avoid volatility (volatility of the memory), SSD manufacturers design the devices with floating gate transistors (FGR) that hold the electrical charge. This allows an SSD to store stored data even when it is not connected to a power source. Each FGR contains a single bit of data, designated either 1 for a charged cell or 0 if the cell has no electrical charge.
Each data block is accessible at a consistent rate. However, SSDs can only write to empty blocks. To work around this problem, SSDs can use overprovisioning, wear leveling, or garbage collection methods. However, the performance of SSDs can slow down over time. Wear levelling balances the workload for Flash cells, while garbage collection deletes obsolete files in the background of operation.
SSDs use four major types of memory, namely single, multi, triple, and quadruple-level cells. Single-level cells can hold one bit of data at a time – one or zero. Single-Level Cells (SLC) are the most expensive form of SSD, but also the fastest and most durable. Multi-Level Cells (MLC) can hold two bits of data per cell and have a larger amount of memory in the same amount of physical space as SLC. However, MLCs have a slower write speed. Three-level cells (TLC) can hold three bits of data in one cell. TLCs have a lower price, but slower write speeds and less durability. TLC-based SSDs offer more flash capacity and are cheaper than an MLC or SLC, but with a higher probability of bit red because they have eight states within the cell. Quadruple-Level Cells can hold four datbist, but again, the lifetime is much shorter than SLCs for example.
Semiconductor manufacturers continue to develop smaller and smaller chipsets that enable high-density SSDs. Intel, Micron, Samsung and Western Digital offer SSDs based on 64-layer NAND flash. Currently, Korean flash manufacturer SK Hynix has claimed the densest SSD – a 72-layer 256GB 3D NAND device. In 2018 Intel and Micron introduced NAND with QLC. Toshiba stacks its SSDs with up to 96 layers.
SSD History and the Introduction to Enterprise Storage
The earliest solid state drives were generally designed for consumer devices. The debut of the Apple iPod in 2005 was the first notable flash-based device to largely penetrate the consumer market.
EMC – now known as Dell EMC – is credited with being the first vendor to integrate SSDs into enterprise storage hardware when it added the technology to its Symmetrix disk arrays in 2008. This led to the introduction of hybrid flash arrays that combine flash drives and HDDs. In most cases, enterprise SSDs in hybrid arrays are used to cache reads in flash. This is due to the higher cost and shorter lifespan of SSDs compared to HDDs.
The earliest commercially developed SSDs were manufactured with enterprise multi-level cell (enterprise MLC) flash technology, which has improved write cycles compared to consumer MLCs. Newer enterprise SSDs are marketed that use Triple-Level Cell (TLC). SSDs manufactured with 3D NAND are the next evolution. IBM, Samsung, Intel and Toshiba have produced and marketed SSDs with 3D NAND, in which flash memory cells are stacked in vertical layers. Toshiba has sold its flash chip business in 2017.
The introduction of flash into businesses is increasing due to improved wear resistance and falling flash prices, although the price decline has been halted by the shortage of global flash supply. Experts claim that SSDs are beginning to displace the traditional hard drive in some applications, although flash drives and HDDs are likely to coexist in many companies in the foreseeable future. For example, SSDs are focused on high performance storage, but less on long-term archiving and backups, where hard drives are typically used.
Applications for Solid State Drives
SSDs offer faster storage and other performance benefits than hard drives. Companies with a rapidly growing need for higher I/O have driven the development and adoption of SSDs. Because SSDs offer lower latency than HDDs, they can efficiently handle both high read and random workloads. The lower latency is due to the ability of a Flash SSD to read data directly and instantly from a specific cell position.
An All Flash Array uses only SSDs as storage. A hybrid flash array combines hard disk storage and SSDs, using the flash to cache hot data that is later written to disk or tape. In server-side flash configurations, SSDs are installed in x86 computers to support targeted workloads, sometimes in conjunction with network storage.
High-performance servers, laptops, desktops or any other application that needs to deliver real-time or near real-time information can benefit from solid-state drive technology. The characteristics of enterprise SSDs make them ideal for offloading reads from transaction-intensive databases, reducing boot storms in virtual desktop infrastructures (VDI), or within a storage array to provide hot data locally for off-site storage in a hybrid cloud scenario.
SSDs are used in a wide range of consumer devices, including computer games, digital cameras, digital music players, laptops, PCs, smartphones, tablets and USB drives. These devices are not designed to provide the same performance or durability as an enterprise SSD.
Several features characterize the design of an SSD. Because it uses no moving parts, an SSD is not subject to the mechanical failure that occurs with HDDs. It is also quieter and uses less power than the HDD. And since SSDs weigh less than hard drives, they are well suited for laptops and mobile computing devices.
In addition, the SSD controller software includes predictive analytics that alert the user to potential drive failure. Because flash memory is customizable, manufacturers of all-flash arrays can manipulate the usable storage capacity with data reduction techniques.
SSD lifetime compared to hard drives
A number of factors affect the lifespan of SSDs and HDDs, including humidity and the effect of oxidizing metals inside the drives. Data recording on both types of media deteriorates over time, with HDDs generally supporting a higher number of write operations per day.
As mentioned earlier, the moving parts of HDDs increase the likelihood of failure. To compensate for this, HDD manufacturers install shock sensors to protect the drives and other components in the devices. This type of sensor uses pressure resistors (piezo resistors) to detect if the device is about to crash and then takes action to shut down the HDD and associated critical hardware.
Heat exposure is another factor that affects drive life, especially for SSDs. Industry experts recommend that unused SSDs be stored at low temperatures to extend their life. Running an SSD at high temperatures for an extended period of time can cause electrons to leak from the NAND flash memory.
Flash supports a limited number of write operations per day. The level of data storage decreases as more and more data is written to the flash cells. Enterprise SSDs are designed with greater endurance than consumer SSDs.
SSD form factors
SSDs do not have the physical limitations of HDDs. This allows drive manufacturers to offer SSDs in different form factors. The most common form factor is a 2.5-inch SSD, available in a variety of heights and supporting Serial Attached SCSI (SAS), Serial Advanced Technology Attachment (SATA) and NVMe (Non-volatile Memory Express) protocols.
The Solid State Storage Initiative (SSSI), a project of the Storage Networking Industry Association (SNIA), has identified three key SSD form factors for the company.
- SSDs that are available in traditional HDD form factors and fit into the same SAS and SATA slots of a server.
- Solid-state cards that use standard add-in card form factors, such as those that use a Peripheral Component Interconnect Express (PCIe) card with a serial port that resides on a circuit board. An SSD with PCIe connectivity does not require network host bus adapters (HBAs) to route commands, which speeds up storage performance. These devices include U.2 SSDs, which are generally considered to be the potential replacement for the miniSATA drives currently used in thin laptops.
- Solid-state modules, which are located in a Dual In Line Memory Module (DIMM) or Small Outline DIMM (SO-DIMM) and can use a standard hard drive interface such as SATA. These devices are known as non-volatile DIMM cards (NVDIMM).
Two types of random access memory (RAM) are used in a computer system: dynamic RAM (DRAM), which loses data in the event of a power failure, and static RAM (SRAM). NVDIMMs provide the permanent memory that a computer needs to recover data. A NVDIMM places the flash memory near the main board, but operations are performed in the DRAM. The flash component is inserted into a memory bus for backup to high-performance memory.
Both SSDs and RAM contain solid-state chips, but the two types of memory function differently within a computer system. As mentioned above, the flash memory is a storage device, while RAM is an active memory that performs the calculations on the data retrieved from the memory.
There are two newer form factors that deserve a mention: M.2- and U.2-SSDs. An M.2 SSD varies in length – typically between 42 and 110 millimetres – and is connected directly to a main board. Communication is via NVMe or SATA. Unlike a traditional SSD form factor, an M.2 device is not hot-swappable, and its small size limits the surface area for heat dissipation, which over time reduces its performance and stability. In enterprise storage, M.2 SSDs are often used as a boot device. In consumer devices, such as a notebook computer, an M.2 SSD provides capacity expansion.
A U.2 SSD describes a 2.5-inch PCIe SSD. These small form factor devices were formerly known as SFF-8639. The U.2 interface allows NVMe-based PCIe SSDs to be inserted into the backplane of a computer at high speed without shutting down the server and memory.
The following video explains the wide range of available form factors for solid state storage and some of the advantages and disadvantages of each form factor.
Nonvolatile SSD memory types
NAND and NOR differ in the type of logic gate used, with NAND devices using an eight-pin serial access to data. NOR flash memory is commonly used in mobile phones, and NOR devices support 1-byte random access.
Compared to NAND, NOR flash offers fast read times, but is generally a more expensive memory technology. NOR writes data in large blocks, which means it takes longer for NOR to erase data and write new data. The random access capabilities of NOR are used to execute code, while NAND flash is intended for storage. Most smartphones support both types of flash memory, with NOR being used to boot the operating system and removable NAND cards to expand the memory capacity of the device.
SSD purchase considerations
Most All Flash Arrays are available as empty chassis or fully populated, allowing customers to choose their preferred SSD to expand capacity or customize configuration. Several generally accepted factors deserve to be considered when purchasing the most suitable SSD.
Data Protection and Error Correction Code (ECC). NAND flash includes tools to detect bit errors and repair reversed bits. As a rule of thumb, ECC requirements increase with the number of cell levels.
The durability. Each SSD warranty covers a finite number of drive cycles, determined by the type of NAND flash. An SSD that is read-only does not require the same durability as an SSD that is intended to be used for most write operations.
Form factor. As mentioned earlier, the form factor determines whether a replacement SSD will work with the existing memory and affects density – the number of SSDs that will fit in a single enclosure – and whether servers will need to be taken offline to replace the SSDs.
Interface considerations. SSDs communicate with a computer processor via an electrical signal. The interface determines the maximum throughput, minimum latency thresholds, and expansion capabilities of the SSD. Manufacturers qualify their SSDs for NVMe, SAS, and SATA, with the SATA interface typically being the most cost-effective. Vendor qualification is designed to help buyers compare devices in terms of capacity, endurance, performance, physical size and price.
Monitoring and management tools. NVME, SAS and SATA use Self-Monitoring, Analysis and Reporting Technology (SMART) to perform health checks to ensure stable performance. SMART monitoring includes automatic alerts and shelf life reports, firmware updates, size changes, SSD formatting, and cleanup operations, among other things.
Power consumption. The drive interface also indicates the maximum performance of an SSD, although many enterprise SSDs are designed to be tuned during operation. This feature allows users to intelligently optimize the performance or power of the device.
Redundant power supply. SSDs contain a small RAM cache to protect transactional and other mission-critical data. Data is stored in RAM and then written to a recently erased flash block on the SSD. This ensures that no data is lost. In addition, enterprise SSDs contain multiple onboard capacitors to power the SSD and ensure that writes from RAM are completed.
For consumer products, the performance benefits of replacing an SSD with an HDD in a laptop are probably not worth the price premium, except for use cases such as high-frequency trading or large-scale PC gaming.
The following video outlines some of the ideal workloads for SSD devices.
The SSD market is dominated by a handful of large manufacturers, including Intel, Kingston Technology, Micron, SK Hynix, Samsung, SanDisk, Seagate Technology, Viking Technology Kioxia (formerly Toshiba Memory) and Western Digital Corp. Micron, Samsung and Seagate produce and sell NAND flash chipsets to solid state drive vendors and also market branded SSDs based on their own flash chips.
The storage capacity of the earliest SSDs was limited compared to the older HDDs. More recently, SSD manufacturers have changed this ratio with new and high-capacity flash drives. Intel, Micron, Samsung and Western Digital offer SSDs based on 64-layer NAND flash.
In 2018, the former manufacturer of all-flash arrays, Nimbus Data, introduced a 100 TB SSD. Korean flash manufacturer SK Hynix has claimed the densest SSD – a 72-layer 256GB 3D NAND device.
Among the established SSD manufacturers, Samsung and Seagate are in a duel to see who wins the competition for SSD capacity. The Samsung PM1643 SSDs from Samsung offer 30 TByte capacity in a 2.5-inch form factor. Seagate has introduced a preview of a 60TB enterprise SSD.
In the past, the prices for SSDs were much higher than those for conventional hard drives. As a result of improvements in manufacturing technology and increased chip capacity, SSD prices have fallen, allowing consumers and enterprise customers to reassess SSDs as viable alternatives to traditional storage media. This phenomenon has been reversed several times in recent years.
The market price for SSDs is influenced by Moore’s Law as well as supply and demand. The development of a dense 3D NAND SSD requires further steps compared to the 2D NAND process. Manufacturers have made efforts to increase yields to meet global demand, with mixed results in recent years.
Between 2015 and 2017, global demand for flash chips exceeded supply. As a result, SSD manufacturers had to hurry to fill their pipelines. Fluctuating demand for flash chips has kept SSD prices variable, but the price of an SSD is still higher than that of an HDD.
A June 2018 report by TrendForce, a research company based in Taipei, Taiwan, said that contract prices began to fall due to an oversupply of flash chips. The resulting drop in prices helped to increase acceptance of SSDs, including PCIe drives.
SSD compared to HDD
The performance of SSDs is considered much faster than that of the most powerful electromechanical drives. Search time and latency are also significantly reduced, and users typically benefit from much faster boot times.
A solid-state drive uses wear leveling to increase drive life. Wear leveling is typically managed by the flash controller, which uses an algorithm to arrange the data so that the write/erase cycles are evenly distributed across all blocks in the device. Another technique is overprovisioning of SSDs to minimize the effects of garbage collection write gain. This limits the usable storage space of the SSD to a certain percentage.
In addition, SSDs have a fixed life expectancy with a limited number of write cycles before performance becomes irregular. This in itself is not a real drawback, as HDDs also wear out over time and eventually fail.
The reading performance of an HDD can suffer if the data is split into different sectors on the disk. The way to repair the disk is with a technique known as defragmentation. SSDs do not store data magnetically, so read performance remains constant regardless of where the data is stored on the drive. Because of their lower latency, SSDs are optimized to perform inline data reduction with minimal impact on application performance.
SSD compared to eMMC
An embedded MultiMediaCard (eMMC) provides the onboard flash memory in a computer. An eMMC is installed directly on the computer’s motherboard. The architecture, formalized by the JEDEC industry group, includes NAND flash and an integrated circuit controller.
An eMMC device has fewer logic gates than an SSD and delivers performance similar to an SSD. The difference is in capacity: a standard eMMC ranges from 32GB to 128GB and is therefore not capable of handling a larger memory footprint.
In portable devices, an eMMC serves as primary memory or as a supplement to removable Secure Digital (SD) cards and microSD multimedia cards. Although this is the historical use of eMMC devices, they are increasingly used in sensors within connected Internet of Things (IoT) devices.
Other types of flash cards for consumer electronics include SD cards for encrypting data on digital devices, removable microSD cards for mobile phones, Secure Digital High Capacity (SDHC) cards for high-resolution images and video, memory sticks for transferring photos and video files, and Plug and Play Universal Serial Bus Cards that plug into a computer’s USB slot.
SSD vs. Hybrid Hard Disk Drive
Although not as widespread as a standard SSD, hybrid hard drives (HHD) can be an alternative that bridges the gap between flash and hard drive magnetic storage. HHDs are used as a way to upgrade laptops, both in terms of capacity and performance.
HHDs have a conventional disk architecture that adds approximately 8 GB of NAND flash as a buffer for disk-based workloads. The HHD controller chip determines whether data is placed on the disk or the SSD module.
Therefore, an HHD is best suited for computers with a limited number of applications, for example to speed up boot times. The price of a hybrid HDD is slightly lower than that of an HDD. In comparison, an SSD is much more expensive due to the integration of more expensive NAND chips.
The advent of All Flash storage arrays
Nimbus Data, Pure Storage, Texas Memory Systems, and Violin Memory were among the start-ups that pioneered the introduction of all-flash arrays, which use solid-state drive memory to replace the hard drive. The success of all-flash start-ups led established vendors to sell upgraded all-flash versions of their traditional disk-based arrays. IBM is considered the first major storage vendor to launch a dedicated allflash array platform called FlashSystem, based on the technology from the acquisition of Texas Memory Systems in 2012.
Also in 2012, EMC acquired XtremIO and has since shipped an all flash system based on XtremIO technology. Other Dell EMC all flash arrays include the flagship PowerMax system, SC Series (formerly Compellent) and PS Series (formerly EqualLogic) systems.
Hewlett Packard Enterprise (HPE) sells 3PAR arrays, all equipped with a flash system, as well as Nimble Storage All Flash and Hybrid SANs. NetApp all flash arrays include the company’s flagship All Flash Fabric-Attached Systems and SolidFire arrays acquired in 2015.
Dell EMC, HPE, Kaminario, Pure, and SolidFire (now part of NetApp) ship All-Flash systems that include SSDs with TLC NAND non-volatile drives.
PCIe SSDs and NVMe Flash devices
Solid-state flash drives have traditionally been designed to use the SATA interface to connect storage to networked servers using host HBAs and other components. A more recent version of server-based flash memory includes SSDs designed to be installed in PCIe slots in servers. Each PCIe-enabled SSD communicates directly with a server mainboard via a dedicated point-to-point connection, essentially eliminating resource contention and reducing latency.
SSD vendors are also developing PCIe devices around the emerging NVMe protocol, a set of specifications designed to operate at the host controller level. NVMe specifications ostensibly aim to increase PCIe device throughput by streamlining the I/O stack and eliminating the latency associated with SAS and SATA-based SSDs.
Applications communicate directly with an NVMe SSD over the PCIe bus. The next expected phase is to develop an ecosystem of NVMe OVER FABRICS (NVME-OF) that enables the transfer of commands between a host server and solid-state storage over Fibre Channel, InfiniBand, and Ethernet.
SSD, PCIe or DIMM Flash: How to choose the right server SSD
When selecting a caching solution, you must select the physical flash device to be used with the caching software. After all, it’s the combination of hardware and software that makes an offering. There are three basic options for server-side flash today:
- Solid-state drives: SSDs are the path of least resistance, but they offer the lowest performance. SSDs are flash devices that come in the form factor of a traditional hard drive. They are connected via SATA or SAS and provide a very cost-effective first step into the solid state world. For many environments, the performance increase provided by a SATA or SAS SSD is sufficient.
- PCIe-based Flash: PCIe-based Flash is the next step in performance enhancement. Although these devices typically offer greater throughput and more IOPS, their real appeal lies in their significantly lower latency. There are also other advantages, including the fact that they do not consume drive bays. The downside is that most of these offerings require a custom driver and only have limited built-in data protection
- Flash DIMMs: Flash DIMMs are another step in reducing latency and go even further than PCIe flash cards by eliminating potential PCIe bus conflicts. However, like PCIe flash cards, they require custom drivers and, uniquely for Flash DIMMS, specific changes to the base I/O system of the read-only memory on the motherboard.
Hybrid DRAM Flash Memory
Advances in SSD manufacturing and other improvements position the technology to play a greater role in non-volatile storage. However, new memory configurations are emerging that combine Flash and server DRAM. These hybrid flash memory devices are a response to DRAM approaching its theoretical scaling limit.
Another type of server-based memory provisioning involves inserting flash memory into DIMM slots on the motherboard. A Flash-based DIMM, also known as in-memory memory, does not need to traverse the PCIe controller or compete with other cards, further reducing latency compared to PCIe Flash cards. Once considered a leader in Flash DIMM cards, Diablo Technologies introduced its ULLtraDIMM and eXFlash DIMM chips through Original Equipment Manufacturer (OEM) partnerships with IBM and SanDisk. Since Diablo was unable to maintain its business model, it filed for bankruptcy in 2017.
Micron and Intel jointly developed permanent memories – under the 3D XPoint brand – which are said to be almost as fast as DRAM, but are priced between DRAM and NAND. The first commercial product based on 3D XPoint is the Intel Optane SSD family.
Prior to its acquisition by Western Digital, SanDisk announced a partnership with HPE in 2015 to develop ReRam, also known as Resistive RAM. The vendors described ReRam as similar to 3D NAND flash, but cheaper to manufacture than DRAM. However, prototypes from the joint project have not yet been realized, and there have been reports that HPE put the project on hold following its takeover by SanDisk.
Similar ReRam initiatives involving Fujitsu and Panasonic are said to be underway, and Crossbar is trying to use ReRam technology to gain a foothold in a number of industries, including the emerging Internet of Things.