480 GB 2.5-inch SATA mixed-use SSD, ordering code MZ7L3480HBLT-00B7C
| Capacity | 480 GB |
|---|---|
| Interface | SATA |
| Form Factor | 2.5-inch 7mm |
| PCIe Version | Not applicable |
| NAND Type | TLC |
| Endurance (DWPD) | 3 |
| TBW | 2,628 TB |
| Sequential Read | 560 MB/s |
| Sequential Write | 530 MB/s |
| Random Read IOPS | 97K IOPS |
| Random Write IOPS | 60K IOPS |
| Power | 2.3 W |
| DRAM Cache | Yes |
| Power Loss Protection | Yes |
| Encryption | Opal 2.0 |
| Condition | New · Used |
Samsung PM897 480 GB, 2.5-inch 7 mm SATA, catalogue code MZ7L3480HBLT, ordering code MZ7L3480HBLT-00B7C. 560 MB/s sequential read, 530 MB/s sequential write, 97K random read and 60K random write IOPS at 4 KB, 2,628 TBW — 3.0 drive-writes-per-day over five years — on Samsung V6 128-layer 3-bit TLC V-NAND, with TCG Opal 2.0 and capacitor-backed power-loss protection. Stocked new or as tested pulls with a 1-year warranty from us.
This is the drive that took the endurance argument off the table for TLC. The SM883 at the same capacity reached 3.0 DWPD and 2,628 TBW only by using 2-bit MLC flash; the PM897 reaches the identical 2,628 TBW and the identical 3.0 DWPD on 3-bit TLC, and does it while moving random write from 27K to 60K IOPS and cutting active power. What it gives back is IOPS consistency, and that trade is the whole subject of this page.
Put the two drives side by side and the shape of the change is clear. Both are 480 GB, both are 2.5-inch 7 mm SATA, both are rated 3.0 drive-writes-per-day for five years, and both therefore carry exactly 2,628 TBW. One is built on 2-bit MLC and one on 128-layer 3-bit TLC. The TLC drive is 20 MB/s faster on sequential read, 10 MB/s faster on sequential write, and more than twice as fast on random write, and it draws 2.3 W instead of 2.8 W on read.
Samsung does not publish the provisioning ratio or the write-amplification figure behind that result, so the mechanism is not documented and this page will not invent one. What is established is the outcome, and the outcome is that at this capacity the choice between MLC and TLC is no longer a choice about endurance. A buyer specifying 3.0 DWPD when the SM883 was current had to accept MLC pricing and MLC supply; the PM897 meets the same figure on TLC with headroom on throughput.
The one number that moves the other way is IOPS consistency: 95% read and 90% write on this drive against 99% and 97% on the SM883. That looks like a regression and, read as a percentage, it is one. Read as a floor it is not, and the next section does that arithmetic, because the figure that constrains a queue is the IOPS the drive actually holds, not the fraction of a larger number it holds.
| Product family | Samsung PM897 (V-NAND SSD, data centre / mixed-use) |
|---|---|
| PN | MZ7L3480HBLT (catalogue code) |
| Ordering code | MZ7L3480HBLT-00B7C |
| Capacity | 480 GB (512-byte sectors, decimal capacity) |
| Form factor | 2.5-inch, 7.00 mm (100.20 × 69.85 × 7.00 mm, max 70 g) |
| Interface | SATA 6 Gb/s, Serial ATA 3.3, ATA/ATAPI ACS-4 |
| NAND | Samsung V6 V-NAND, 128-layer, 3-bit (TLC) |
| Sequential read / write | 560 MB/s / 530 MB/s (128 KB, QD32) |
| Random 4 KB (QD32) | 97K IOPS read / 60K IOPS write |
| Random 8 KB (QD32) | 57K IOPS read / 40K IOPS write |
| IOPS consistency | 95% read / 90% write at 4 KB, sustained |
| Latency and QoS | 105 µs read / 30 µs write at QD1; 99.99th percentile 0.2 ms read, 0.05 ms write at QD1; 0.6 ms and 0.7 ms at QD32 |
| Endurance | 2,628 TBW — 3.0 DWPD over five years (TBW = DWPD × 365 × 5 × capacity) |
| Power | Active read 2.3 W; active write 3.0 W; idle 1.4 W (DIPM off); supply 5 V ±5% |
| Power-loss protection | Yes — capacitor hold-up transfers cached DRAM data to flash on failure; no hold-up time published |
| Data security | AES-XTS 256-bit encryption engine, TCG Opal 2.0 compliant |
| Features | TRIM, NCQ to depth 32, S.M.A.R.T., end-to-end data integrity, RoHS |
| Reliability | MTBF 2.0 million hours; UBER 1 sector per 1017 bits read |
| 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 (non-operating) |
| Code | Type | Condition | Quote |
|---|---|---|---|
| MZ7L3480HBLT | Catalogue code | New / used | Enquire |
| MZ7L3480HBLT-00B7C | Ordering code | New / used | Enquire |
The -00B7C suffix is specific to this build and is not interchangeable with the -00A07 or -00B07 codes, which belong to two other parts at the same byte count: the read-intensive PM893a and the PM897a build of this same drive. Those carry different endurance ratings or a different firmware identity, so a PO written against one will not be filled with the other. Quote the full string and we will ship against exactly that.
The part-number table also records which build goes with which capacity, and it is the only place the pattern is documented: the 480 GB and 960 GB members carry HBLT, while the 1.92 TB and 3.84 TB members carry HBNA. The 960 GB member is the same drive at twice the bytes and twice the write budget, with identical performance figures. The range stops at 3.84 TB in each direction — there is no 240 GB PM897 and no 7.68 TB PM897, so a deployment that starts on this drive and grows past 3.84 TB per device has to change family, not just capacity.
60K IOPS of 4 KB random write is 240 MB/s of small-block write traffic, which is a substantial fraction of what a SATA 6 Gb/s link can carry at all. It is the highest random-write figure Samsung publishes for any 480 GB SATA drive in this line-up: the PM893 at 480 GB is rated 29K, the older PM883 at 480 GB 24K, and the current-generation PM893a at 480 GB 17K. At the same 480 GB and on the same 128-layer TLC, that spread is a firmware and provisioning decision rather than a silicon one.
The queue-depth figure shows up in the tail latency, which is where a queue actually feels it. At QD1 the two drives are close — 0.2 ms read and 0.05 ms write here against 0.2 ms and 0.1 ms on the SM883. At QD32 the write tail separates: 0.7 ms on this drive against 1.9 ms on the SM883, for the same 99.99th percentile. That is what a database waiting on a commit, or a hypervisor draining a write-back cache, is actually measuring, and it is the number that justifies the drive in a journal or cache role rather than in a boot role. The read side gives up a little in exchange: 105 µs read latency against 85 µs on the MLC part, and 95% read consistency against 99%. Neither matters on a volume whose reads are cached upstream, and both matter on a read-serving tier — which is the one role this drive should not be given.
2,628 TB over five years is 1.44 TB of writes every day. Written flat out at 530 MB/s the rating would last about 57 days, so the ceiling is real, but nothing that lives on 480 GB writes at 530 MB/s for two months. The daily allowance is the useful figure, and 1.44 TB covers a busy journal, a write-back cache, a continuously rewritten metrics store, or a mirrored boot pair with room to spare.
Now the consistency arithmetic, because the percentages are misleading on their own. 95% of 97K read is 92K IOPS held in the sustained state; 99% of 97K on the SM883 is 96K, so the read floor is about 4% lower. On the write side, 90% of 60K is 54K IOPS held, against 97% of 27K on the SM883, which is 26K. The write floor is therefore more than twice as high on the drive with the worse percentage. A queue that needs 40K sustained write IOPS fits on this drive and does not fit on the one that reports 97%.
This is why the honest summary of the generation change is not “TLC is worse” or “TLC is better”. The average write rate doubled, the sustained write floor doubled, the endurance held, power fell, and read latency and read consistency each gave up a little. Which of those dominates is a property of the workload, not of the drive.
480 GB at 3.0 DWPD is a mixed-use specification, and the roles that fit it are the ones that write continuously at a rate a read-intensive part would not survive: a database journal or WAL volume, a write-back cache in front of object storage or a slower tier, a logging or metrics store rewritten every day, a ZFS special or log vdev pair, a hypervisor host's local swap and boot devices. In each of those the 60K random-write figure and the 0.7 ms QD32 write tail are the numbers that decide whether the role works.
Three things are worth checking before it goes on a BOM. Power-loss protection is present and capacitor-backed, and Samsung's brochure states that cached DRAM is transferred to flash when a failure is detected, but no hold-up time is published — so it covers a brief interruption and does not replace a battery-backed controller cache or a UPS in a design that needs a guaranteed flush window. TCG Opal 2.0 is claimed, so self-encrypting deployments and remote locking are supported, unlike the General SED parts Samsung shipped in the previous generation. And the family ceiling is 3.84 TB per drive, which for dense arrays means more devices per unit of capacity than an NVMe design would need.
If the workload is mostly reads, the trade reverses and the read-intensive parts are the better buy: the PM893 at 480 GB gives 98K random read and Opal 2.0 at 876 TBW and 1.0 DWPD for a lower price per drive. This drive costs more per byte written because it is built to be written to.
| Model | Capacity | Sequential read / write | Random 4K read / write | Endurance | Class |
|---|---|---|---|---|---|
| PM897 480 GB | 480 GB | 560 / 530 MB/s | 97K / 60K IOPS | 2,628 TB (3.0 DWPD) | Mixed-use |
| SM883 480 GB | 480 GB | 540 / 520 MB/s | 97K / 27K IOPS | 2,628 TB (3.0 DWPD) | Mixed-use, MLC |
| SM863a 480 GB | 480 GB | 510 / 485 MB/s | 95K / 19K IOPS | 3,080 TB (3.5 DWPD) | Mixed-use, MLC |
| PM893 480 GB | 480 GB | 550 / 520 MB/s | 98K / 29K IOPS | 876 TB (1.0 DWPD) | Read-intensive |
| PM893a 480 GB | 480 GB | 560 / 360 MB/s | 97K / 17K IOPS | 876 TB (1.0 DWPD) | Read-intensive |
| PM883 480 GB | 480 GB | 550 / 520 MB/s | 98K / 24K IOPS | 683 TB (0.78 DWPD) | Read-intensive |
| PM863a 480 GB | 480 GB | 520 / 480 MB/s | 97K / 16K IOPS | 683 TB (0.78 DWPD) | Read-intensive |
Two rows carry the argument. Against the SM883 this drive is the same endurance on cheaper flash with 2.2 times the random write, so for a write-heavy role there is no reason left to pay for MLC. Against the PM893a — the current-generation read-intensive part at the same 480 GB — the difference is almost entirely write: 530 against 360 MB/s sequential and 60K against 17K random, for three times the endurance. Those two drives share a NAND generation and a form factor; what separates them is what each is provisioned to do.
The one row that beats this drive on endurance is the SM863a at 3.5 DWPD and 3,080 TBW, and it does so at 510/485 MB/s and 19K random write on MLC. If nothing but the largest possible write budget will do, that is the part; if the workload runs a queue rather than a steady trickle, this one is more than twice as fast where it counts.
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 batch. Send the quantity, the condition you need and the platform the drives are going into, and you will get a quote with lead time the same working day. Matched pairs for mirroring are selected 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