HARD DRIVES & SSDs
Since you’re on the topic of PCs/laptops for now, let’s add in about storage.
There are basically two types of storage that you’ll find in devices - HDD (hard disk drive) and SSD (solid state drive).
We won’t get technical about some of those built on and such, just to touch on the two main technologies.
HDD
Regular hard drives contain moving parts. You’ll hear terms like head, platter, spindle, read/write head and others. This type of drive physically spins a disk (platter) that stores data magnetically. The information is accessed using the read/write head.
SSD
You can think of an SSD as a large thumb drive. It stores the data in chips - not magnetically. This makes your SSD much faster than your HDD. You could consider “built in” drives on boards as SSDs as well - but with NetBooks these are much smaller in size/capacity. As mentioned above - NetBook drives may be as small as 32GB/64GB.
HYBRID
Some manufacturers have hybrid drives. These are somewhat of a merger of HDD and SSD, so you get the device at a lower price point. Not as expensive as an SSD but not as cheap as a HDD.
FAILURES
There is nothing made by man that doesn’t deteriorate over time. Hard drives have an estimated life span that is measured in work hours, age and total read/write cycles. This is the same for all types of drives. Manufacturers previously used MTBF (mean time between failure) as the benchmark, but now some have moved to AFR (annualized failure rate). It’s basically the same thing in the end - they give you a timeframe that they think the drive will last. Factors affect this such as abuse and improper powering off and such.
Please also note that HDDs and SSDs can both fail without warning, but typically the HDD will give some “signs” of failure. You may have the system slow down in performance for example - with an SSD you will likely not see a speed difference even when the drive is aging and failing. The SSD is more hardy than the HDD when it comes to “user errors” however. You can bump and even drop an SSD while it’s in use, and provided the chips aren’t damaged it will continue working. A HDD that’s dropped however, will likely stop working or give problems right after.
Technology has progressed and SSDs are now more reliable and affordable. If you have an older system and find it running slow - switch to an SSD and you may get an overall performance boost.
When it comes to hard drives, there’s something that I’d like to touch on. Testing, backup and failures.
I recently spoke to a friend learning IT - the method used for testing the system involved checking the storage (hard drive / HDD) last. While I don’t think there’s any “wrong way” to go about doing checks, I wouldn’t do it that way. Here’s why.
Running a basic SMART test on a drive will tell you the general health. From there you can run any other disk checking software you like and periodically monitor the health of the drive. I do this first instead of last - because there are so many different failure points for a storage device. You don’t want to be running extended operations or tests and the drive fails before you have a chance to pull any data for the client. Regardless of what it is, the most important thing to the customer is the data. So here’s my path when I get a system.
- If it turns on - boot to a Linux or WinPE environment.
- Run SMART tests on the drive.
- Run RAM tests.
- Run system stress tests if required.
After those are done, depending on the results I will proceed as needed. If the drive is found bad - immediately turn it off and advise the client to get a replacement drive. Set a disclaimer that if they decide to continue with it, it’s entirely on them. If you have an external storage device with enough space - do a backup for the client. I have a bad habit of holding the customer backup for more than 6 months and forgetting that I have it. I do this in case they get back the system and the drive fails - they’re aware that I did the backup. Sometimes this is an issue, because some clients will take it for granted that you have it - your responsibility - and just continue usage. This is why you need the disclaimer and a timeframe within which they need to have a replacement done.
A RAM test is usually done if there are system instabilities. The stress test is usually for CPU testing.
Having an external boot medium also allows you to do virus/spyware scans if needs be - and allows you to clean the system outside of the OS environment. This becomes a bit harder on some EFI systems where they lock you out of the BIOS, but it’s still doable.
RECORD KEEPING
So I was asked tonight by a friend about a ticketing system. This is something that I believe every tech should use and get into the habit of using. Whether you self-host or use a service - free or paid - you should have something in place to keep track of the work you do, and alert your clients of the progress of the job. The blog link below shows a couple of options that you can use. Pick one and test it out.