1.92 TB M.2 22110 NVMe SSD, PCIe 4.0, ordering codes MZ1L21T9HCLS-00A07 / -00B7C
| Capacity | 1.92 TB |
|---|---|
| Interface | NVMe |
| Form Factor | M.2 22110 |
| PCIe Version | PCIe 4.0 |
| NAND Type | TLC |
| Endurance (DWPD) | 1 |
| TBW | 3,504 TB |
| Sequential Read | 5,500 MB/s |
| Sequential Write | 2,000 MB/s |
| Random Read IOPS | 800K IOPS |
| Random Write IOPS | 85K IOPS |
| Power | 8 W |
| DRAM Cache | Yes |
| Power Loss Protection | Yes |
| Encryption | AES-XTS 256 |
| Condition | New · Used |
Samsung PM9A3 1.92 TB M.2 22110, catalogue code MZ1L21T9HCLS, orderable as -00A07 and -00B7C. 5,500 MB/s sequential read and 2,000 MB/s write on PCIe 4.0 x4, 800K random read and 85K random write IOPS, 3,504 TBW at 1 DWPD. 110 mm long, 22 mm wide, 3.8 mm thick, 128-layer TLC V-NAND, capacitor-backed PLP, AES 256-bit full disk encryption, 8 W active in either direction and 2.5 W idle.
Two terabytes inside a slot that needs no cable is the reason this capacity exists in this carrier. The module gives up more to the U.2 version of the same drive than the 960 GB does — 700 MB/s of sequential write rather than 100 — and the arithmetic behind that gap is the most useful thing on this page, because it tells you which capacities to pair with which chassis.
| Product family | Samsung PM9A3 (PCIe 4.0 data centre NVMe SSD, 1 DWPD) |
|---|---|
| PN | MZ1L21T9HCLS (catalogue code) |
| Ordering codes | MZ1L21T9HCLS-00A07 / MZ1L21T9HCLS-00B7C |
| Capacity | 1.92 TB (unformatted; 1 GB = 1 billion bytes, formatted capacity is less) |
| Form factor | M.2 22110, single-sided, 110.0 × 22.0 × 3.8 mm, about 20 g |
| Interface | PCIe 4.0 x4, NVMe 1.4, M-key |
| NAND | Samsung V6 (128-layer) V-NAND TLC |
| Sequential read | 5,500 MB/s (128 KB transfer size, queue depth 32) |
| Sequential write | 2,000 MB/s |
| Random 4K read | 800K IOPS (queue depth 32) |
| Random 4K write | 85K IOPS (queue depth 32) |
| Endurance (TBW) | 3,504 TB |
| Endurance (DWPD) | 1 drive write per day for five years |
| Power | 8.0 W active read, 8.0 W active write, 2.5 W idle with DIPM off |
| Power-loss protection | Yes, capacitor-backed, with end-to-end data path protection |
| Encryption | AES 256-bit full disk encryption, TCG Opal 2.0, IEEE 1667 encrypted drive |
| MTBF | 2.0 million hours |
| Environment | 0 to 70 °C operating, 5 to 95% non-condensing humidity |
| Features | TRIM, garbage collection, S.M.A.R.T., secure boot |
| Model | Capacity | Carrier | Seq read / write | Random 4 KB | TBW | Power |
|---|---|---|---|---|---|---|
| PM983 | 1.92 TB | M.2 22110 | 3,000 / 1,400 MB/s | 480K / 42K | 2,733 TB | 8 W |
| PM9A3 | 960 GB | M.2 22110 | 5,000 / 1,400 MB/s | 550K / 60K | 1,752 TB | 7.5 W |
| PM9A3 | 1.92 TB | M.2 22110 | 5,500 / 2,000 MB/s | 800K / 85K | 3,504 TB | 8 W |
| PM9A3 | 1.92 TB | U.2 7 mm | 6,800 / 2,700 MB/s | 850K / 130K | 3,504 TB | 10 W |
| PM9D3a | 1.92 TB | U.2 15 mm | 12,000 / 3,500 MB/s | 1,700K / 150K | 3,504 TB | 11.2 W |
The first row is the previous-generation module at the same capacity and the same 8 W: 3,000 and 1,400 MB/s against 5,500 and 2,000 here. That is the gain the Gen4 controller bought in this carrier, and it is close to a straight doubling on sequential and on random read, because both generations were limited by the same thing.
Rows two and three are the two capacities of this series in this carrier, and reading them together is more informative than reading either alone. From 960 GB to 1.92 TB the sequential read moves 500 MB/s, but the sequential write moves 600 and the random write moves 25K, while write power rises by only 1.5 W against 4.5 W for the same step in the U.2 carrier. Doubling the flash bought forty-three percent more write, not a hundred, because the module had much less headroom to spend on it.
Compare the two carriers at both capacities and the pattern is visible without needing a theory for it. At 960 GB the module gives up 100 MB/s of sequential write against its U.2 twin, about seven percent. At 1.92 TB it gives up 700 MB/s, or twenty-six percent. Random write tells the same story: 60K against 70K at the small capacity, 85K against 130K here.
The corresponding power figures explain the direction. The U.2 drive goes from 8.0 W of active write at 960 GB to 12.5 W at 1.92 TB, a 4.5 W increase, and its write bandwidth rises by eighty percent. The module goes from 6.5 W to 8.0 W, a 1.5 W increase, and its write bandwidth rises by forty-three percent. A 110 mm module has a power ceiling that a 2.5-inch shell with a 12 V rail does not, and write bandwidth is what gets spent against it.
Practically: this module is at its most competitive in the middle of its range, and at its least competitive at the top. The 3.84 TB version quotes 2,000 MB/s write as well, on the same 8.2 W, so going up one more capacity buys endurance and nothing else.
The natural role is a pair of modules in a 1U chassis running a redundant system volume. Two of these are 3.84 TB of mirrored boot capacity and 7,008 TB of combined endurance, taking no front bays and no backplane, at 16 W under load and 5 W at rest. A hypervisor plus its images and logs fits comfortably, and the front bays stay free for the data drives that have to be serviceable from outside the chassis.
The second role is local cache inside a node that already has network storage, where the requirement is capacity near the CPU rather than capacity in the rack. Here the 2,000 MB/s write matters more than the 5,500 read, because a cache is written to as often as it is read from, and it is the figure that separates this module from the 960 GB version.
What it is not is a substitute for a U.2 drive in a system that has U.2 bays. The same money buys 6,800 MB/s read and 2,700 MB/s write in a carrier that can be swapped from the front, and no amount of convenience in the slot makes up for opening a chassis during an incident.
One drive write a day for five years on 1.92 TB is 1.92 TB of host writes a day, or 22 MB/s averaged. At the rated 2,000 MB/s a full fill takes about sixteen minutes, so the daily budget covers a complete rewrite. At the random write ceiling of 85K IOPS the module absorbs 348 MB/s, and 3,504 TB at that rate lasts about a hundred and sixteen days. That is a wider margin than the U.2 version's seventy-six, not a narrower one: the module burns the same rating more slowly precisely because its write ceiling is lower, so the lower ceiling is a working limit rather than an endurance advantage.
| Code | Type | Capacity | Condition | Quote |
|---|---|---|---|---|
| MZ1L21T9HCLS | Catalogue code | 1.92 TB | New / used | Enquire |
| MZ1L21T9HCLS-00A07 | Ordering code | 1.92 TB | New / used | Enquire |
| MZ1L21T9HCLS-00B7C | Ordering code | 1.92 TB | New / used | Enquire |
Two suffixes in this carrier, both describing the same module. The prefix is the part that has to be read correctly, because MZ1L is the M.2 22110 drive in this family, MZQL is the 2.5-inch U.2 drive at the same capacity and a higher rating, and MZTL is E1.S. All three are sold as PM9A3 1.92 TB and only one of them fits a given slot.
All stock, factory sealed or pulled and tested, comes with one year of cover from us, and it is recorded against the serial so it stays with the drive if it moves to another machine. Pulls are erased, graded and photographed before packing, and the SMART log is filed against the serial rather than against the order. Delivery is three to seven working days from dispatch, with tracking issued when the parcel leaves. The minimum is two pieces per line, other capacities in this family can travel on the same order, and pricing is returned the same working day once part number, condition and quantity are confirmed.
Tell us who you are — we reply within one working day with a firm quote, stock status and lead time. Volume discounts available.
15.36 TB E1.L 1U long ruler NVMe SSD, PCIe 4.0, ordering code MZEL215THBLA-00AMV
7.68 TB E1.S 1T NVMe SSD, PCIe 4.0, ordering codes MZCL27T6HBLA-00AMV / MZTL27T6HBLA-00A07
3.84 TB E1.S 1T NVMe SSD, PCIe 4.0, ordering codes MZCL23T8HCLS-00AMV / MZTL23T8HCLS-00A07
1.92 TB E1.S 1T (9.5 mm) NVMe SSD, PCIe 4.0, ordering code MZCL21T9HCJR-00AMV