1.92 TB 2.5-inch SATA mixed-use SSD, ordering code MZ7KH1T9HAJR-00005
| Capacity | 1.92 TB |
|---|---|
| Interface | SATA |
| Form Factor | 2.5-inch 7mm |
| PCIe Version | Not applicable |
| NAND Type | MLC |
| Endurance (DWPD) | 3 |
| TBW | 10,512 TB |
| Sequential Read | 540 MB/s |
| Sequential Write | 520 MB/s |
| Random Read IOPS | 97K IOPS |
| Random Write IOPS | 29K IOPS |
| Power | 2.8 W |
| DRAM Cache | Yes |
| Power Loss Protection | Yes |
| Encryption | SED |
| Condition | New · Used |
Samsung SM883 1.92 TB, 2.5-inch 7 mm SATA, catalogue code MZ7KH1T9HAJR, ordering code MZ7KH1T9HAJR-00005. 540 MB/s sequential read, 520 MB/s sequential write, 97K random read and 29K random write IOPS, 10,512 TBW — 3.0 drive-writes-per-day over five years — on 2-bit MLC V-NAND behind a Samsung Maru controller. Stocked new or as tested pulls with a 1-year warranty from us.
Nothing about the performance changes between 960 GB and this drive. Both run 540/520 MB/s, both reach 29K random write and 14K at 8 KB, both hold 99% read and 97% write consistency. The only figure that moves is the write budget, which doubles to 10,512 TB. That makes this a capacity purchase rather than a performance one, and it is worth being explicit about it: if 960 GB is enough room, the smaller drive delivers the same speed for less money; if it is not, nothing above 960 GB will make the array faster, only bigger.
What the capacity does change is the shape of the deployment. One 1.92 TB drive replaces two 960 GB drives, which halves the number of write streams the host has to schedule, the number of firmware revisions to track, and the number of devices that can fail independently. Against a mirrored pair of 960 GB drives the arithmetic is neat: the pair carries 5,256 TB of budget each and gives you 960 GB of usable space with redundancy, while this single drive gives 1.92 TB of usable space with the same 10,512 TB of total budget — twice the room for the same write allowance, and no mirror to rebuild.
The rest of the package is the generation's: 2-bit MLC V-NAND in 2 Tb two-chip-enable packages (the JR code in the model name), a documented power-loss sequence where capacitor hold-up flushes the DRAM cache to NAND and the mapping table is rebuilt within ten seconds, 99.99th-percentile latency of 0.2 ms read and 0.1 ms write at queue depth 1, a 2-million-hour MTBF, and no TCG Opal. It is a 2018 design, and it is out of production.
At the top of the performance curve and in the middle of the capacity ladder. Random write peaks at 29K IOPS at 960 GB and stays there; the capacity above this one, 3.84 TB, runs the identical 540/520 and 97K/29K, and only its endurance differs. So the ladder above 960 GB is a ladder of bytes and write budget, and this rung doubles both.
Endurance at this capacity is 10,512 TB, which is 5.76 TB of writes every day for five years. The number stops being a constraint here and becomes a property of the deployment: a database volume, its log and a staging copy can share the device without anyone modelling write amplification first. That is the practical difference between this drive and the read-intensive parts at the same capacity, which sit an order of magnitude lower.
The trade against the previous generation does not narrow at this capacity. The SM863a at 1.92 TB carries 12,320 TBW at 3.5 DWPD — 17% more write budget, exactly the ratio it holds at 240 GB and 960 GB, because both ranges scale endurance linearly with capacity. What the 883 gives back is speed: 540/520 against 510/485, and 29K against 28K random write.
| Product family | Samsung SM883 (V-NAND SSD, data centre / mixed-use) |
|---|---|
| PN | MZ7KH1T9HAJR (catalogue code) |
| Ordering code | MZ7KH1T9HAJR-00005 (General SED) |
| Capacity | 1.92 TB (3,750,748,848 sectors of 512 bytes) |
| Form factor | 2.5-inch, 7.0 mm (100.20 × 69.85 × 6.80 mm, max 60 g) |
| Interface | SATA 6 Gb/s, ATA/ATAPI-7 and Serial ATA 3.2 |
| Controller | Samsung Maru (Samsung LSI), with DRAM write cache |
| NAND | Samsung V-NAND, 2-bit MLC (2 Tb dies, two-chip-enable packages) |
| Sequential read / write | 540 MB/s / 520 MB/s (128 KB, QD32) |
| Random 4 KB (QD32) | 97K IOPS read / 29K IOPS write |
| Random 8 KB (QD32) | 57K IOPS read / 14K IOPS write |
| IOPS consistency | 99% read / 97% write at 4 KB, sustained |
| Latency and QoS | 85 µs read / 35 µs write at QD1; 99.99th percentile 0.2 ms read, 0.1 ms write at QD1; 0.6 ms and 1.9 ms at QD32 |
| Endurance | 10,512 TBW — 3.0 DWPD over five years (Samsung: TBW = DWPD × 365 × 5 × capacity) |
| Power | Active read 2.8 W; active write 3.7 W; idle 1.4 W; supply 5 V ±5% |
| Power-loss protection | Yes — capacitor hold-up flushes DRAM to NAND; mapping table rebuilt within 10 s (SPOR). No hold-up time published |
| Data security | AES 256-bit encryption engine (General SED part); no TCG Opal claim |
| Features | TRIM, NCQ to depth 32, S.M.A.R.T., end-to-end data protection, RoHS |
| Reliability | MTBF 2.0 million hours; UBER 1 sector per 1017 bits read; data retention 3 months at 40 °C after rated endurance |
| Environmental | Operating 0 °C to 70 °C (case); non-operating −40 °C to 85 °C; humidity 5–95% non-condensing |
| Shock and vibration | 1,500 G at 0.5 ms half sine; 20–2,000 Hz at 20 G peak, 7–800 Hz at 2.17 Grms |
| Status | End of life — Samsung has discontinued the SM883 |
| Code | Type | Condition | Quote |
|---|---|---|---|
| MZ7KH1T9HAJR | Catalogue code | New / used | Enquire |
| MZ7KH1T9HAJR-00005 | Ordering code | New / used | Enquire |
The model name reads the same way as every other capacity in the family: MZ for memory, 7 for the 2.5-inch 7 mm SATA form factor, K for the VM line-up (the 2-bit MLC V-NAND family), H for the Maru controller, 1T9 for 1.92 TB, H for the BGA NAND package, A for second-generation flash, JR for the 2 Tb 2-CE NAND density, and -00005 for the General SED customer code. The suffix is a security and packing designation rather than a capacity variant, and it is the field a PO and a warranty claim are written against.
5.76 TB of writes every day for five years, or 3.0 drive-writes-per-day at 1.92 TB. Written flat out at 520 MB/s the rating would be gone in about 234 days, so a genuinely scratch-heavy workload is still capable of exhausting it — but nothing that lives on a 1.92 TB SATA device writes at 520 MB/s for eight months. The realistic reading is that endurance has stopped being the limiting design constraint at this capacity.
The figure reconciles exactly with Samsung's published formula: 3.0 × 365 × 5 × 1,920 GB = 10,512 TB. On the 863a generation the brochure drive-writes number and the TBW number disagreed by a tenth of a drive-write, which left buyers guessing which one the warranty would be measured against; on this generation there is one number and it is reproducible.
Two useful comparisons. A mirrored pair of 960 GB drives holds 5,256 TB each, so the pair owns the same 10,512 TB of write budget as this single drive while presenting half the usable capacity — the mirror buys availability with bytes, not with endurance. And the read-intensive PM893 at this capacity carries 3,504 TBW, exactly a third, which is the cleanest statement of what the mixed-use rating is worth at 1.92 TB.
This is the first capacity in the range where the drive can be the data volume rather than the system volume: a VMFS or LVM datastore for ten to twenty guests, a small database with its log on the same device, an analytics or build workspace that is rewritten continuously, backup staging that is written once and read during restores, a ZFS mirror pair carrying an appliance's state. In each case the capacity removes the pressure that makes a 240 GB or 480 GB drive a compromise, and the endurance removes the need to model writes before specifying it.
The limits are the generation's, unchanged by capacity. Power-loss protection is capacitor-based and documented but unquantified — it covers a short interruption, not a sustained outage, so a battery-backed controller cache or a UPS still belongs in the design. There is no TCG Opal 2.0 claim: the -00005 part is a General SED device with the AES 256-bit engine, which encrypts at rest but gives no Opal locking or remote management. And the drive is out of production, so the supply is remaining stock, and at 1.92 TB the economics of buying a matched pair while stock exists are worth thinking about before a rebuild forces the question.
Where current-generation credentials matter more than write behaviour, the PM893 at 1.92 TB is the read-intensive alternative with Opal 2.0 and a published PLP specification at 3,504 TBW. It is the right drive for a mostly-read datastore and the wrong one for a workspace that is rewritten all day. The honest summary for this capacity: buy the SM883 when the volume is written continuously and the room is needed; buy the PM893 when it is mostly read and a third of the write budget is already generous.
| Model | Capacity | Sequential read / write | Random 4K read / write | Endurance | Class |
|---|---|---|---|---|---|
| SM883 480 GB | 480 GB | 540 / 520 MB/s | 97K / 27K IOPS | 2,628 TB (3.0 DWPD) | Mixed-use |
| SM883 960 GB | 960 GB | 540 / 520 MB/s | 97K / 29K IOPS | 5,256 TB (3.0 DWPD) | Mixed-use |
| SM883 1.92 TB | 1.92 TB | 540 / 520 MB/s | 97K / 29K IOPS | 10,512 TB (3.0 DWPD) | Mixed-use |
| SM883 3.84 TB | 3.84 TB | 540 / 520 MB/s | 97K / 29K IOPS | 21,024 TB (3.0 DWPD) | Mixed-use |
| SM863a 1.92 TB | 1.92 TB | 510 / 485 MB/s | 95K / 28K IOPS | 12,320 TB (3.5 DWPD) | Mixed-use |
| PM863a 1.92 TB | 1.92 TB | 520 / 480 MB/s | 97K / 24K IOPS | 2,733 TB (0.78 DWPD) | Read-intensive |
| PM893 1.92 TB | 1.92 TB | 550 / 520 MB/s | 98K / 30K IOPS | 3,504 TB (1.0 DWPD) | Read-intensive |
Read the top four rows together and the shape of the range is clear: throughput and random write stop improving at 960 GB, and from there only capacity and write budget grow, doubling at each step to 21,024 TB at 3.84 TB. This capacity is therefore the middle of the ladder — the rung to take when 960 GB is too tight and 3.84 TB is more room than the deployment can use.
Against the SM863a of the same capacity the trade is the familiar one, and at this capacity it is at its narrowest: 540/520 against 510/485 and 29K against 28K random write, for 10,512 TB against 12,320 TB of endurance. The older drive keeps a 17% bigger write budget; the newer one wins on throughput and consistency. Against the read-intensive pair the gap is wide enough to decide the purchase on its own: the PM893 reads faster and writes 30K random, but holds exactly a third of the endurance, and the PM863a a quarter of it.
Every drive carries a 1-year warranty from us against defects in workmanship and function. Verified faulty units are replaced or refunded and we cover return shipping. Stock is held in both new and used condition; the condition you order is confirmed before dispatch, and used units are pull-tested rather than untested lots. Stock is held in our own warehouse and normally dispatches within three to seven working days, with tracking issued at dispatch.
This capacity is quoted rather than listed, because price moves with condition, quantity and which batch the units come from — the SM883 is end-of-life, so larger quantities are quoted against confirmed stock rather than a standing lead time. Send the quantity, the condition you need and the server model the drives are going into, and you will get a quote with lead time the same working day. Pairs destined for a mirror are matched from a single batch where stock allows, including firmware revision and wear level.
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