1

After about 2 months of trying to fix my questions, contributing and finally very short conversation with one of moderators I got rid of question ban.

Moderator that helped me suggested that I should put my next question on meta and see people feedback before I put my question on superuser.

I have question related with SSD.


Title: How can I measure how much data I write daily?

I would like to run few experiments and see how much data is being written on my disk when I do specyfic tasks (programming, running virtual machines, working with databases, audio editing etc.).

I want to use that data to estimate SSD lifespan and decide how big SSD I really need to avoid killing SSD in 1 year. I have read somewhere that average user writes 20GB/day and that value may be used to estimate SSD lifespan, however I am probably not average user.

I know that Windows Performance Monitor exists and there is "Average bytes / Write" counter or something very similar (im not using non-english Windows), but... is this really good idea to count written bytes?


Can I post it without being banned again risk of being downvoted hard?

  • "Can I post it without being banned again?" is a little hard to answer, no? If your question is whether it's "a good idea", then I would encourage you to rephrase it into something that can be objectively answered. Why would it not be a good idea? I mean.. after all you are trying to get the bytes written. And when you say "Average bytes / write counter or something" you may want to clarify what that something is, show a screenshot, etc. Be as clear as possible. – slhck Nov 19 '14 at 15:41
  • I wrote "or something" because I'm not using English version of Windows. I edited my question. – Kamil Nov 19 '14 at 15:44
  • Looks ok to me, although people could question the premise of such a post. I mean, you cannot even reliably estimate the lifespan of such a device because it may fail anytime. – slhck Nov 19 '14 at 15:59
  • @slhck I know that SSD will not die after writing exactly 1000x240GB but I want to replace SSD when it reach 1000x240GB. I know thats just estimation. If these people will question my idea - It would be off-topic, because I want to ask how much data I write, not if this is good method to estimate SSD lifespan... – Kamil Nov 19 '14 at 16:42
  • Right, but you are making assumptions that this is a good method, which may be questioned, or may be interpreted as an X/Y problem — basically, what allquixotic says below. – slhck Nov 19 '14 at 17:00
  • From Intel SSD 520 Series product specyfication: "Minimum Useful Life/Endurance Rating - 5 years. The SSD will have a minimum of five years of useful life under typical client workloads with up to 20 GB of host writes per day." This is industrial standard to measure SSD endurance (JESD218). – Kamil Nov 19 '14 at 17:16
  • 1
    Most of the time a SSD will come with an OEM's tool to report the amount of data written to the device. It also will report the health of the time and very the device is writing and reading data at the appropriate speeds. I personally don't find a question helpful asking how to keep track of the daily writes, when that specific data point isn't all that important, considering only the lifetime writes with regards to SSD is important. – Ramhound Nov 20 '14 at 12:24
  • Thanks @Ramhound, but I wanted to do it on system where is no SSD yet. – Kamil Nov 20 '14 at 12:26
3

In my opinion, your original question that you state at the beginning starts out great, but then you water it down by adding things like "or something" and then conclude by asking if it's "a good idea".

Never ask "is X a good idea" on a StackExchange site. This is always a matter of personal opinion and your question will be closed.

Your question should almost always start with "What" or "How". You can then explain exactly the criteria of what you're looking for if you want to make your question longer; you can also tell us what you've already tried in your attempts to solve the problem yourself.

Questions that start with "Why" or in any way are asking for opinions, can easily be closed. Not all questions of this form are closed, but you really have to know what you are doing in order to ask this type of question. Therefore, to minimize the risk of your question being downvoted/closed, you should avoid asking questions that solicit opinions until you have a firm footing on the site.

I'd rewrite your question as follows. Note that this may still be considered to be an X/Y problem: someone may disagree with you that measuring your daily disk write amount is a useful way of determining how quickly you'll kill your SSD. However, that is an argument over a technical point, and they could probably answer your question by telling you that they disagree with this premise, and then suggest what you can do instead.

At least it wouldn't be closed as off-topic or opinion-based, after rewriting it as below.

Some general guidelines to consider:

  • When you are looking to accomplish some task using software, ALWAYS state the exact operating system and version you are using. Even if this information is not strictly required for the question to be answered, it is still helpful for the answerers to understand what kind of environment you're running on.

  • Be self-aware of what you are asking. Re-read your question several times, and look for anything that you could possibly have a reason to doubt. For example, are you 100% positive that your methodology will help you reliably determine how long your SSD will last? If you are not sure, you should revise your question: rather than assuming something that you are not sure of, you should ask the more general question that would resolve your doubt there. It is much harder to write a good answer to a question if it contains false statements or unsubstantiated claims. By only stating what you are sure of, and asking in your question what you are not sure of, you minimize the risk of being told you are suffering from the X/Y problem.

  • Keep working on your English. I know it is not your native language and you are trying, but keep trying. :) You will get better!


Title: How can I measure how much data I write daily?

I would like to run a few experiments and see how much data is being written on my disk when I perform specific tasks (e.g., programming, running virtual machines, working with databases, audio editing etc.).

I believe that measuring my usage on a day to day basis will be helpful, because I can remember what I did during that day, so if I observe my disk writes on a daily basis, I can reflect on what I might have done during that day that took so little (or so many) bytes written. I can then consider the expected MTBF bytes written of the SSD as stated by the manufacturer, and compare that to how many bytes I write per day based on which activities I perform, and calculate an estimate for how many days my SSD will last before it would, on average, probably start to fail.

I want to use that data to estimate SSD lifespan and decide how big of an SSD I really need to avoid killing the SSD in 1 year, since larger SSDs can level out the write wear across more NAND cells, so they tend to last longer. I have read somewhere that the average user writes 20 GB/day and that value may be used to estimate SSD lifespan, however I am probably not an average user. My usage may be much less or dramatically more than this; that is part of the reason I want to measure this value.

How can I gather this data on Windows INSERT VERSION HERE?

  • Thanks for very constructive answer. However I still think my method of lifespan estimation is OK. This is actually covered by JESD218 industrial standard: "SOLID STATE DRIVE (SSD) REQUIREMENTS AND ENDURANCE TEST METHOD". – Kamil Nov 19 '14 at 17:20
2

Kamil, I'm not familiar with your other questions that apparently led to a ban. Allquixotic provided a great answer about asking questions here. However, let me suggest another perspective. The problem may not be entirely that you are asking bad questions, you may be asking them in the wrong place.

An analogy. The super-hardware store where I shop has snacks and beverages (because guys can get hungry when they get lost at a hardware store). The supermarket where I shop has a small section of an aisle with basic hardware items. However, when I need hardware, I go to the hardware store; when I need food, I go to the food store. You go to where you are likely to find what you need.

Support sites are like that. If they have survived the test of time, they have a core of people with some expertise. You can probably go to any site and find someone there who can answer your question, regardless of what it is. However, each site is somewhat specialized. Some are focused on a narrow range of subject matter. Super User was set up to handle certain kinds of questions and it does a great job with questions that are within its defined scope. However, you are swimming against the current trying to force-fit questions on this site that are fundamentally outside of what the site specializes in.

That includes things like "big" questions, where a good answer would require a text book. If the answer to a question is likely to start with "it depends", or "there are many possibilities depending on...", that is a warning flag that the scope may be too broad. Another is if you need to pick the brains of a lot of people with experience in an area (looking for opinion shaped by experience more than facts that can be looked up). Questions that involve "brainstorming" ("help me conceptualize some ways to accomplish this task") are not as good a fit as "is this procedure known to work?".

It's a continuum. Every question is different, but the farther a question gets from the "core" of the site, the more chance there is of downvotes or closure.

You have the questions you have, and there is nothing wrong with the questions, themselves (I'm not talking about how you word them, just what you are pondering in your head). If a question is not a good fit at this site, it can be a tortuous task to shoe horn it into a form that won't be downvoted or closed. In general, you may have better success at collecting information or getting the kind of answer you need by asking it on a site that normally handles those kinds of questions. Part of preparing each question should be asking yourself, "Where is the best place to ask this?"

  • I agree with this answer. This website is about people asking questions that solve an actual problem (I.e. "Windows will not boot"). Its very difficult to ask a good question that is not based on trying to solve a problem. An exception of a good question, that does not have a specific problem, is perhaps an all encompassing question on how to remove malware – Ramhound Nov 20 '14 at 12:30

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .