Back to Blog
Education

SSD or HDD? The Cost-Per-Terabyte Decision Nobody Frames Correctly

The SSD-versus-HDD debate is usually argued on speed. That is the wrong frame. It is a cost-per-terabyte decision crossed with a workload decision -- and once you frame it that way, the answer stops being a matter of opinion.

Mark | | 12 min read
SSDHDDStoragePrice Per TBNANDCommodity PricingBuying GuideUnit Economics

The SSD-versus-HDD question gets argued endlessly, almost always on the wrong axis. One side says SSDs are faster; the other says HDDs are cheaper; both are correct and neither has actually answered the question. Because the real decision is not “which is better.” It is “which delivers the terabytes I need at the lowest cost, given the speed my workload actually requires.”

Framed that way, the debate dissolves into arithmetic. And the arithmetic starts, as it always does with storage, at cost per terabyte.


Two technologies, two very different cost curves

Solid-state drives (SSDs) store data in NAND flash — no moving parts, extremely fast access, and a cost per terabyte that has fallen dramatically over the years but remains well above spinning disk. Hard disk drives (HDDs) store data on magnetic platters — mechanical, slower, and by far the cheapest terabytes available at scale.

The persistent structural fact: HDDs cost meaningfully less per terabyte than SSDs, and that gap widens as capacity grows. For bulk storage — archives, backups, media libraries, anything measured in double-digit terabytes — spinning disk is dramatically cheaper. For performance-critical storage — an operating system drive, a database, active project files, anything where access speed is the bottleneck — SSDs earn their premium by delivering a kind of performance HDDs simply cannot.

So the two technologies do not really compete. They occupy different points on a cost-versus-speed plane, and the correct choice depends entirely on where your workload sits on that plane.


The framing that resolves it: tier your storage

The professional answer to SSD-versus-HDD is not to choose one. It is to tier: put each kind of data on the storage whose cost-per-terabyte-and-speed profile matches that data’s needs.

  • Hot data — the operating system, applications, active databases, files you are working on right now — goes on SSD. Here you are paying for speed, and the speed pays for itself in responsiveness and throughput. You buy relatively few terabytes, so the higher $/TB is affordable.
  • Cold data — backups, archives, finished projects, media you access occasionally — goes on HDD. Here speed is nearly irrelevant; you want the cheapest terabytes you can get, and spinning disk delivers them. You buy many terabytes, so the lower $/TB compounds into real savings.

This is why serious storage setups mix both, and why the “which is better” framing is a category error. A fast, small SSD for hot data and a large, cheap HDD for cold data is not a compromise — it is the cost-optimal answer for almost every real workload. You pay the SSD premium only on the terabytes that need it, and the HDD discount on everything else.


Running the numbers honestly

To make a tiering decision, compute cost per terabyte for each tier and each candidate, then apply the speed filter:

RoleData typeRight technologySort bySpeed requirement
OS / appshotSSD (NVMe)$/TB within SSDshigh — non-negotiable
Active projectswarmSSD$/TB within SSDshigh
Backups / archivecoldHDD (CMR)$/TB within HDDslow — irrelevant
Media librarycoldHDD$/TB within HDDslow-moderate

Within each tier you still apply the full storage discipline: rank by cost per terabyte, filter for the constraints that make terabytes non-interchangeable (for HDDs, CMR versus SMR; for SSDs, DRAM cache and endurance rating), and choose the cheapest terabytes that clear the bar. The tiering decision tells you which technology; the unit price tells you which specific drive. For the cold tier, our hard drive $/TB comparison already ranks the cheapest spinning terabytes across every retailer we track.

A subtle point on SSDs: their cost per terabyte is not the whole story, because they also have an endurance limit — a finite amount of data that can be written over their life. For write-heavy workloads, the honest metric extends toward cost per terabyte-written. For the typical hot-data role, endurance is rarely the binding constraint, but it is worth checking on drives destined for heavy, sustained writes.


How the supply cycle changes the calculus

Both NAND (SSDs) and HDDs ride supply cycles, and they do not always move together. When a demand shock hits NAND particularly hard, SSD cost per terabyte can spike while HDD holds steadier — or vice versa. This can temporarily widen or narrow the gap between the two technologies and shift the tiering math at the margins.

The discipline absorbs this without changing. You are already computing cost per terabyte for each tier; when the cycle moves, the numbers move, and you re-sort. If SSDs spike, the case for putting warm-but-not-hot data on HDD strengthens. If HDDs tighten, you might tolerate a bit more on SSD for the convenience of a single fast tier. The framework tells you what to recompute; the current unit prices tell you what to conclude. Tracking the current $/TB on spinning disk hour by hour is the fastest way to see when the cycle has moved the HDD tier under you.

Mark’s Take: The SSD-versus-HDD debate is a fake debate, kept alive by arguing speed against price as if they were the same axis. They are not. Storage is a cost-per-terabyte decision crossed with a speed-requirement decision, and once you separate those two axes, the answer is almost always “both, tiered” — SSD for the terabytes that need speed, HDD for the terabytes that need to be cheap. Compute $/TB within each tier, filter for CMR and endurance, and let the workload decide which tier each byte belongs to. The people still arguing which is “better” are answering a question that was never the real one.


The bottom line

SSDs and HDDs are not competitors to be ranked; they are tools for different jobs on the same cost-versus-speed plane. Hot data wants SSD speed and can afford the higher cost per terabyte because there is not much of it. Cold data wants HDD cheapness and does not care about speed. The optimal system uses both, and the cost-per-terabyte discipline — applied within each tier — tells you exactly which drive to buy for each role.

Stop asking which is better. Ask which terabytes need speed and which need to be cheap, put each kind of data where it belongs, and pay the premium only where it earns its keep.


MarketCrystal provides trend analysis and market commentary for informational purposes only. Nothing in this publication constitutes financial advice or purchasing recommendations. Storage prices are cyclical and change constantly; always verify current pricing before buying. Past pricing does not guarantee future results.


Follow the plumbing.

Enjoyed this article? Get early access to Trismegistus.

The boards are free forever. Trismegistus — our analysis engine, built on frontier AI — is coming. Join the first 100 subscribers for free founding access when it launches, plus the market read in your inbox.

MarketCrystal

An oracle for markets — the real price, in clear view.

Coverage

  • Stablecoins & Digital Money
  • CBDC & Monetary Policy
  • Blockchain Infrastructure
  • AI Infrastructure
  • Semiconductors & Memory
  • Energy Systems

About

MarketCrystal is an independent market price-analysis engine. We normalize every market to a real cost per unit -- $/TB, $/GB VRAM, $/W, $/Wh, $/gal -- across commodities, hardware, energy, and digital-money data, so the true price is always in clear view. Our AI analyst, Mark -- powered by the Trismegistus engine -- reads what those prices mean.

© 2026 MarketCrystal. All rights reserved.

Analysis only. Not financial advice.

No ads. No tracking. Rankings never sold. Why?