5

In this question on Super User they say that Solid-State Drives (SSDs) use much less power than traditional Hard Disk Drives (HDDs). I read some internet posts and research articles that backup this claim, although some people say the difference is less than you expect.

My question is: does this mean that overall SSDs have a lower environmental impact over their entire lifespan (so including manufacturing and disposal)? Have there been any scientific studies on this?

With environmental impact I mean the aspects that are generally part of a life-cycle analysis so energy use, water use, GHG emissions, use of rare earth metal and toxicity. Preferably all aspects relevant for disk drives, but if there's only data on a single aspect like carbon footprint I'm interested as well.

4
  • 1
    I note you don't consider recycling, which is accurate but also horrible. Especially the magnets in legacy disks which contain accessible rare earths. And note that a lot of hard disks are used in airconditioned or at least force cooled evironments, so the electricity saving is not just the label difference. Lifetime is harder to calculate and changes all the time, but I suspect SSDs are close to legacy disks now. – Móż Mar 11 at 1:52
  • @Móż actually with disposal I meant the end-of-life of the product in a more general sense, whatever it may be. I know a lot of electronics are dumped in lower income countries, so that's probably the most common scenario. However, I'm open to studies that assume recycling. – THelper Mar 11 at 14:17
  • @THelper You need a rule for environmental impact that reduces to a single figure of merit. How do I compare something that uses 50kWh/GB vs 200 kWh/GB over total lifespan but uses 200 gal of water vs 10,000 gal of water to make? vs 10 pounds/GB of low problem industrial waste vs 1 pound of highly toxic industrial waste? This is a problem common to all sustanabilty questions. – Sherwood Botsford Mar 15 at 18:09
  • 2
    @SherwoodBotsford I disagree that a single figure is needed. Saying that product A has a lower carbon footprint but higher water footprint than product B would be a perfectly fine answer. Such results are very common in LCAs. I do grant you that calculating all aspects yourself would be way too much for an answer, so I'll make clear that I'm actually looking for results from research studies. – THelper Mar 15 at 18:47
0

SSDs store the data in semiconductors. HDDs store the data in rotating rust platters.

In a large scale, if you only need to access the data rarely, HDD has a competitive edge, especially if the access is so rare you can spin down and up the disks as needed (so that the constantly spinning disk doesn't consume energy). The reason is that semiconductor production is a relatively dirty and you need lots of semiconductor chips to store any meaningful amounts of data. So for backup purposes HDDs win.

However, if the data storage needs are small, and/or if the access to the data is frequent, SSDs will probably win. The reason being that for storing only small amounts of data (such as in a single phone) needs only one small semiconductor chip, and frequent access to a data with HDDs would require a massive RAID 1 mirror of the data so that all accesses can be satisfied at a great rate. So you would need to duplicate the data perhaps ten perhaps hundred times with HDDs if access rates are great, but only once if you are using SSDs that are many orders of magnitude faster than HDDs.

3
  • What do you mean by "rust platters"? – RockPaperLz- Mask it or Casket Apr 10 at 13:25
  • 1
    @RockPaperLz-MaskitorCasket, one of the materials that's been used for the magnetic layer of hard drives is iron oxide, aka rust. Most hard drives these days use cobalt alloys, but the term "spinning rust" for hard drives has stuck around. – Mark Apr 12 at 20:26
  • @Mark Thanks for the explanation Mark. – RockPaperLz- Mask it or Casket Apr 12 at 22:37
0

Three factors that are not typically considered in these situations:

  1. The amount of computer usage time saved using an SSD over an HDD.
  2. The amount of industrial equipment built specifically for a single product, as well as the proportion of equipment that can be used for multiple processes.
  3. Where and how is the product being recycled/disposed?

All of these are very difficult to determine carbon footprint values for and can vary wildly between manufacturers. Unfortunately, it would be very unlikely to get a manufacturer to agree to an outside comprehensive study of their processes to understand their environmental impact to extrapolate for a consumer scale.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.