Disc temperature and Malwarebytes.

[Nazareno]

Hello, when I run a scan with Malwarebytes the temperature of my disk goes from 32 degrees to 43 degrees. In spite of having configured the analysis so that it has less priority, it still raises the temperature of the disk. This does not happen with Eset Online Scanner for example. Could you solve it? Thank you. Best regards.

[exile360]

Greetings,

Unfortunately I doubt there is much they can do to alleviate this as they are already executing the process as a low priority task and I don't believe there is any way to programmatically reduce disk I/O for the scan engine.  It is likely that just based on the way that it reads file, folder and other data structures that it causes a level of disk usage resulting in the temp increase.  However, with that said, 43 is not very hot at all and should do no harm to the disk.  In fact, if it is an SSD that you're talking about, the NAND flash memory chips actually prefer to run at higher temperatures as this enhances their performance/speed, and actually running them too cold can result in failure and data corruption.  It is the controller chip that you must be concerned with when it comes to SSDs (particularly NVMe SSDs) as they do tend to throttle when they get too warm, reducing the disk's performance (though it doesn't harm the disk or its lifespan, it simply slows performance temporarily to allow the chip to cool down, similar to how a CPU or GPU throttles when the temps get too high), however the typical throttling temperature for a modern NVMe SSD is around 70~80 degrees Celsius, so there should be no issues if it is only getting as warm as 43C.

[Nazareno]

Thank you very much for your reply. I enclose the disc usage of some of your competitors. I think you can improve it, if you propose it. Best regards.

 

 

 

 

 

[exile360]

Yes, I'm sure it is possible to some extent, however I'm just as certain that the scan engine would lose a lot of speed if they did so, and scan speed/short scan times are and have always been one of the highest priorities for Malwarebytes, not only for performance, but also because the faster a scan can be completed, the more quickly any present threats can be eliminated before they are able to do further damage to the system.

With that said, I will pass on your feedback to the Product team for consideration.

[Nazareno]

Thank you very much. If you need something more or some data about the system environment where I run Malwarebytes to be able to replicate this just tell me and I will gladly help you in whatever I can. I thank you again. Best regards.

[David H. Lipman]

I'll assume that's 43 Deg Celsius which is ~110 Deg Fahrenheit.

That's not a high temperature.  Obviously if a disk is being scanned, it is in its highest mode of operation ( SSD or Spindle ) and its temperature will be expected to rise above a quiescent state temperature.  Your highest mode of operation temperature is both reasonable and within operation specifications.

Comparatively speaking Humans are ~98.6 Deg F.  Your disk drive temperature isn't much higher.

The application priority is a moot point.  The disk activity is the key to drive temperature and any application that causes numerous files to be open and read successively will create a highest mode condition of operation for a given disk drive.

 

Edited November 12, 2019 by David H. Lipman
Edited for content, clarity, spelling and grammar

[Nazareno]

Maybe it's a problem not Malwarebytes but the brand of WD disks, I currently have a ssd of that brand. I used to have a Kingston ssd and I don't remember the record reaching those temperatures.

[nukecad]

I think it's a trade off - speed of scan against temprature the drive reaches.

For an extreme comparison think of it like a car.

You can drive it at 50 Mph and the engine is going to be at low/normal temps. Go 100 Mph and the engine temp is going to rise. Go 150 Mph and you are pushing it.

How quick you want to get to the end of the journey is your choice of vehicle (scanner) and how quickly you want it to get to the destination (scan finished).

Edited November 12, 2019 by nukecad

[exile360]

Just as an additional point of reference, I have a laptop and I use a Samsung 960 Pro 1TB NVMe PCIe SSD and it has two temperature sensors built in which can be read using HWINFO64.  The first, which represents the temperature of the NAND (remember, NAND flash memory LIKES to be HOT because it performs better at higher temperatures and is more reliable when run that way; if you wan't to kill an SSD quickly, run it at sub-ambient temps; it will die real quick) is currently sitting at 43C at idle (nothing but my browser and a couple tabs open, no active scans etc.) and the second, which represents the temperature of the controller (THIS is the component you want to keep cool, not so much because it will fail, but because it will throttle, reducing the drive's speed/performance temporarily until it cools down) is currently sitting at 63C which is approximately 7 degrees below its throttle point (Samsung SSD's use micro controllers that throttle around 70C), and while it can certainly reach this temperature easily under a heavy load (like performing a scan), it doesn't actually do any harm to the drive (this is the entire reason this throttling functionality exists in these drives to begin with).

Also keep in mind that as far as I know, this doesn't even apply to normal SATA SSDs at all, only M.2/NVMe drives because normal SSDs aren't fast enough to come anywhere near generating the kind of heat that could harm their micro controllers (again, it's the controller chips that don't like heat, NOT the NAND flash storage; the NAND likes to operate under high temps and it's actually better for it than colder temps).

You can learn more about all this at the following links:

https://www.pugetsystems.com/labs/articles/Samsung-950-Pro-M-2-Throttling-Analysis-776/
https://www.pugetsystems.com/labs/articles/Samsung-960-Pro-vs-950-Pro-Performance-Thermal-Throttling-868/
https://edgeup.asus.com/2016/m-2-ssd-throttling-tests-vindicate-asus-motherboard-designs/

It is much more the temperature of the controller than anything else that determines how high temperatures will impact a flash storage device, and NAND flash can endure much higher temperatures than the micro-controllers built into them can which is why it is the controller's temp that determines when/if the drive throttles.  Unless your drive reaches extreme temperatures (well above 100C), there is no danger to the NAND itself as it is optimal for it to run at higher temps.

I keep hammering this point because it is a subject I've done a LOT of research on for the past several years, ever since I got my first NVMe storage drive in a laptop (like you, I was nervous about the heat situation which is why I did all this research in the first place and how I now know how the drives work, and what does and doesn't pose a real risk to the drives' health with regards to temperatures).

[Nazareno]

The temperature of the disk is a symptom, the problem is the intensive use of the disk by Malwarebytes even configuring the analysis with low priority.

[David H. Lipman]

The temperature of the disk is a by-product of normal operation and is within normal operating specifications.  A Western Digital Spindle drive Operating Environmental condition is 0 ~ 60 Deg C and you reported 43 Deg C.

There is no issue.

[Nazareno]

I've already detected what the problem is. In the custom scan the Rootkits option makes the use of the disk rise but also the temperature. Without the option to scan for Rootkits there is no intensive use of the disk. They should improve that. I will continue to test Malwarebytes and make observations. Thank you.

 

[exile360]

Yes, the rootkit scan performs a raw data/sector analysis of the disk.  There is nothing to improve; it is an intensive scan and thus has to place an intensive load on the disk, however as has already been stated multiple times, a rise in 10 degrees C is anything but a hazard to the drive, particularly since the highest temperature it hit was well within the safe operating specifications for the drive itself.  Computer hardware uses power, power generates heat; that's reality.  You cannot possibly have a system where the temperatures of hardware components do not go up during heavy use.  It's the reason they put heatsinks on CPUs and GPUs, heat spreaders on RAM, heatsinks on chipsets and heatpipes and/or heatsinks on VRMs for power delivery on motherboards, because they know that these components generate a lot of heat during operation and so these cooling accessories are added to them in order to help ensure that they remain within safe operating temperatures to prevent damage, and as long as the components in a system do remain within their safe ranges of temperatures there is no problem, at least as far as heat is concerned (current/voltage is another matter, however that is beyond the scope of this conversation as it is unrelated and is connected more to overclocking).

Edited November 15, 2019 by exile360