Data Storage Reviews Storage Storage drives

PCIe 4.0 with a new and cheaper concept: MSI SPATIUM M450 NVMe M.2 SSD 1 TB Review

Phison PS5019-E19T

The package of the relatively new Phison PS5019-E19T controller already attracts attention with its small dimensions. It measures a whole 7 × 11 mm and already suggests a clearly fancy, inner simplification. The casing of the well-known Phison PS5016-E16, for example, is even almost three times larger, although the same 28 nm node from TSMC process is used for manufacturing. The reasons for the shrinkage cure are also quickly explained.

Thus, the Phison PS5019-E19T is based on a single-core rather than dual-core ARM Cortex R5 processor. For this, the number of possible channels for connecting the flash memory is reduced to four instead of eight, which also somewhat limits the use of certain modules (number, size). By the way, the DRAM interface is also missing, but I already wrote something about the unbuffered operation on the first page. The Phison PS5019-E19T is thus a truly extreme (but purposeful) stripped-down controller designed for building low-cost, unbuffered SSDs with PCIe 4.0 support. And it is also more frugal with a power consumption of well under 4 watts, which makes it significantly cooler than the E16 or E12 models.

Abandonment of the DRAM buffer

Buffer-free products like the Samsung 980 or the WD Blue SN550 have been on the market for months and show that it is possible to do without DRAM with certain restrictions. However, the PCIe 4.0 bus with increased bandwidth, which should theoretically let the SSD controller work more efficiently with the PC’s RAM, does not always help. This is because it is also important to remember that the lack of a dedicated DRAM forces the controller to use only HMB (Host Memory Buffer) technology to work with the address translation table.

Unfortunately, this also means that the drive will fork a certain part of the PC’s memory (up to 64 MB) for its own purposes during operation in order to buffer calls to the flash memory array. As a result, the SSD’s speed decreases like an avalanche under high load in certain scenarios, which especially affects the read process very much. If you have to process more than 32 GB of information in one go, it can get quite tight.

The drop in random reading can be more than half, or it can be mitigated or circumvented by tricks in the controller. However, it is a fact that an unbuffered SSD is rather poorly suited for many intensive workloads. You have to scrutinize yourself, the area of application and the applications used. For that, I’ll have a detailed workstation test with the comparison to a “conventional” PCIe 4.0 SSD with DRAM later.

Micron’s 176-layer TLC 3D NAND (codename B47R)

The Phison PS5019-E19T just described is compatible with a wide range of flash memory variants and supports modern NAND chips with 1200 MHz interface. And because it’s so nicely compatible, if you’re a low-cost NVMe SSD vendor, you can also use Micron’s latest 176-layer TLC 3D NAND (codenamed B47R) with it. For example, the MSI Spatium M450 I tested contains two IABHG94AYA chips, each with eight 512-gigabit TLC 3D NAND blocks inside.

By the way, Micron’s still quite new 176-layer flash is the first real in-house development by Micron after the separation from Intel. This memory is also very different in its internal structure from the products of all other manufacturers. The cell structure of the Micron NAND relies on replacement gate, which ends up being a cross between floating gate and charge trap technologies. Micron replaced the polysilicon in the gate material with metal in this NAND and switched to a different process for etching the vertical cylindrical channel, resulting in an increased diameter.

In addition, a non-conductive layer of silicon nitride has been introduced into the structure of the micron memory, in which data can be stored. This layer also reduces quite effectively the mutual interference of the cells in such a vertical stack. Of course, we can go into a little more detail. The next figure shows a chip looking at the metal top, a chip looking at the CMOS circuits (CuA) and the markings of the B47R chip. Compared to the older B37R TLC die, the chip size has been reduced by a whopping 25% due to the higher cell density, larger number of 3D NAND cells and effective scaling of the design. The B47R is now Micron’s second generation CTF structure after the B37R.

Source: TechInsights

The height of the NAND cell array (for comparison: the height from the selector on the source side to the BL) is now more than 11 µm. The total number of gates, including selectors (STs) and dummy word lines (DWLs) per vertical NAND strand is 195, or 195T, which is the highest number ever achieved in 3D NAND. They rely on double-stack architecture, the replacement gate process, charger trap nitride (CTN) and CMOS-under-array (CuA) techniques. The bit density reaches 10.273 Gb/mm2 for the 512Gb TLC die.

Thus, in the end, it was possible to achieve a number of positive effects, because it was possible to increase the speed of the programming cells, increase their stability, reduce power consumption and at the same time also improve the reliability of data storage. Ultimately, Micron’s 176-layer TLC-3D NAND thus takes the next step in 3D flash technology in general. And in terms of performance, Micron claims that the new memory has reduced read and write latency by about 35% compared to the previous generation 96-layer memory. Well…

 

Kommentar

Lade neue Kommentare

M
Massaker

Mitglied

86 Kommentare 27 Likes

Traumhafte Spitzenwerte im mittleren Screenshot von ATTO...allerdings nur beim "Laufwerk D:"... ich vermute MSI M450 war dabei "Laufwerk E:", Guten Morgen!:coffee::giggle:(y)

Antwort 1 Like

O
Oberst

Veteran

339 Kommentare 131 Likes

Klingt nach einem vernünftigen Produkt, hat aber durchaus auch gute Konkurrenz in Form der WD Blue SN570 oder WD Black SN750 SE, die bieten ähnliche Leistung für aktuell unter 100€.
Bezüglich QLC: Würde ich nicht als Sackgasse bezeichnen, das braucht einfach auch noch Zeit. Die TBW steigen da auch. Ich finde aber, das größere Problem für QLC ist eher die Schreibleistung bei vollem Cache.

Antwort Gefällt mir

Homerclon

Mitglied

83 Kommentare 40 Likes

Wie sieht die Leistung unter PCIe 3.0 aus? Wenn man einzig die Übertragungsleistungen betrachtet, dürfte es keinen spürbaren Unterschied machen. Aber Theorie und Praxis ...

Antwort Gefällt mir

c
cunhell

Urgestein

556 Kommentare 523 Likes

Wenn ich mir die Benchmarks ansehe ist z.B. eine WD SN750 PCIe 3.0 ist ähnlich performant und man bekommt die SSD immer wieder unter 100€. Nur weil PCIe 4.0 draufsteht macht das doch das Gesamtprodukt nicht lukrativer. Jeder NVMe PCIe 4.0 Slot kann auch NVMe 3.0.
Man kauft eine PCIe 4.0 SSD und bekommt nur PCIe 3.0 Performance. So eine SSD macht die Performancevorteile von PCIe 4.0 quasi obsolet.

Und der unbedarfte Kunde glaub was Tolles zu kaufen weil 4.0 statt 3.0 drauf steht.

Cunhell

Antwort Gefällt mir

H
Headyman

Veteran

117 Kommentare 54 Likes

Ich sehe das auch mit etwas gemischten Gefühlen, Hauptsache PCIe 4.0 steht drauf :rolleyes:.
Und das die Übertragung auch noch massiv einbrechen soll, wenn die Kiste unter Last ist? Also das ist doch kontraproduktiv, wenn ich viele Daten lesen und verarbeiten will, dann ist doch "immer" was los, und dann wird das noch ausgebremst...
Der Preisvorteil ist ja ganz nett und die Anwendungsperformance ist anscheinend doch ganz gut (im Gegensatz zu den Benchmarks).

Warten wirs mal ab, evtl. sieht man ja in den professionellen Workloads noch Argumente.

Vielleicht wird das ja auch ein neuer Benchmark für DDR5, der kann ja besser mit parallelen Workflows...:LOL:

Antwort Gefällt mir

Igor Wallossek

1

10,276 Kommentare 19,019 Likes

Naja, also der Vergleich zur "normalen" PCIe 4.0 SSD war gar nicht so schlecht. Ich habe die SSD jetzt mal einige Tage als System-SSD genutzt und subjektiv keine Verschlechterung festgestellt. Und ich bin wirklich Power-User :D

Antwort Gefällt mir

S
Salierfähig

Neuling

7 Kommentare 4 Likes

Diese und die SN770 machen doch eine echt gute Figur für SSDs ohne DRAM. Klar, eine Corsair MP600 Pro XT (auch mit 176-Layer Micron) ist da noch mehr ausgeglichen im Gesamtbild, aber die kostet ja auch mehr und schluckt den doppelten Strom.
Der Sprung von PCIe3 ohne DRAM zu PCIe4 ohne DRAM ist ja schon gewaltig. Deshalb frage ich mich: Ist da nochmal so ein großer Sprung für PCIe5 ohne DRAM drin? Und ist DRAM dann ab PCIe6 sogar überflüssig, weil es fast keinen Mehrwert bringt? Das wäre echt wild.

Antwort Gefällt mir

Danke für die Spende



Du fandest, der Beitrag war interessant und möchtest uns unterstützen? Klasse!

Hier erfährst Du, wie: Hier spenden.

Hier kannst Du per PayPal spenden.

About the author

Igor Wallossek

Editor-in-chief and name-giver of igor'sLAB as the content successor of Tom's Hardware Germany, whose license was returned in June 2019 in order to better meet the qualitative demands of web content and challenges of new media such as YouTube with its own channel.

Computer nerd since 1983, audio freak since 1979 and pretty much open to anything with a plug or battery for over 50 years.

Follow Igor:
YouTube Facebook Instagram Twitter

Werbung

Werbung