• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Intel bug incoming? Meltdown and Spectre exploits

Can anyone recomend any free to use benchmarking software for use in a comercial environment? I.e. for me to benchmark our machines before and after patches at work? If i have to buy us something i will but in 10 years this is the first real time i've had to do this kind of testing...
 
Luckily we have SCCM which can push out BIOS updates to our 2000ish desktop PCs. The main problem is all the Hyper-V and VMware hosts.
 
Ok maybe I'm an idiot, maybe not though.
If someone writes a piece of code that manages to reliably extract information from the processor it is run on, but because nobody had ever thought of doing it before, is it suddenly the processor manufacturers fault or the guy who wrote the code.
This seems to be what's happening to INTEL right now, in my view. Just because an exploit has been found in their processors, they have been doing the same thing for the last 10 years or so, but all of a sudden they are to blame for people finding an exploit.
Makes no sense to me.:confused:
 
bru said:
Ok maybe I'm an idiot, maybe not though.
If someone writes a piece of code that manages to reliably extract information from the processor it is run on, but because nobody had ever thought of doing it before, is it suddenly the processor manufacturers fault or the guy who wrote the code.
This seems to be what's happening to INTEL right now, in my view. Just because an exploit has been found in their processors, they have been doing the same thing for the last 10 years or so, but all of a sudden they are to blame for people finding an exploit.
Makes no sense to me.:confused:
Click to expand...
I don't think people are blaming Intel for other people finding an exploit in Intel's faulty CPUs.

What they are blaming Intel for is for failing to spot the fault in their CPUs for ten years, for hushing it up for six months whilst continuing to sell faulty CPUs and for their CEO dumping stock three months after Google told Intel that their CPUs were faulty - OK?
 
KiNgPiN83 said:
Can anyone recomend any free to use benchmarking software for use in a comercial environment? I.e. for me to benchmark our machines before and after patches at work? If i have to buy us something i will but in 10 years this is the first real time i've had to do this kind of testing...
Click to expand...

Benchmarks aren't going to tell you anything useful, the impact is very workload specific. You need Performance/Load testing software.
 
mid_gen said:
Benchmarks aren't going to tell you anything useful, the impact is very workload specific. You need Performance/Load testing software.
Click to expand...

I know what you mean, it was more for a rough guide of how the total available performance would get affected from "new server A with no patch" to "new server A with patch". (Which is obviously quite a vague indicator but an indicator none the less.) I'm leaving the SQL guys to test their stuff as they have far more knowledge on SQL than me. I'm just coming at it from an Infrastructure point of view.
 
OK so apparently my motherboard has an Overclocking profile thing in the BIOS where you can save your BIOS settings to a file on a USB drive. Am I right in thinking that to update the BIOS all I need to do is save the settings to USB, update the BIOS and then load the settings back again? I really hope the update process is easy.
 
bru said:
Ok maybe I'm an idiot, maybe not though.
If someone writes a piece of code that manages to reliably extract information from the processor it is run on, but because nobody had ever thought of doing it before, is it suddenly the processor manufacturers fault or the guy who wrote the code.
This seems to be what's happening to INTEL right now, in my view. Just because an exploit has been found in their processors, they have been doing the same thing for the last 10 years or so, but all of a sudden they are to blame for people finding an exploit.
Makes no sense to me.:confused:
Click to expand...

It's because Intel basically implemented a broken security model in their chips. What they put in there to stop people doing this exact thing can be tricked because it doesn't do it's job properly. The suspicion is that this was done to give Intel an edge in performance. Other manufacturers like AMD are not anywhere as weak as Intel are because their design is better for preventing this ie does what it's supposed to.

It's like all these expensive cars being stolen because thieves have learned to relay the signals from the keyless entry fobs. Is it the car manufacturers fault that this is a relatively easy thing to do with their new keyless entry security systems? Most people would say yes, and Intel is in the same situation ie, what they built for security doesn't work properly.
 
bru said:
Ok maybe I'm an idiot, maybe not though.
If someone writes a piece of code that manages to reliably extract information from the processor it is run on, but because nobody had ever thought of doing it before, is it suddenly the processor manufacturers fault or the guy who wrote the code.
This seems to be what's happening to INTEL right now, in my view. Just because an exploit has been found in their processors, they have been doing the same thing for the last 10 years or so, but all of a sudden they are to blame for people finding an exploit.
Makes no sense to me.:confused:
Click to expand...
They will have knows it was a possibility. They decided on a shortcut. To gain free performance.
 
KiNgPiN83 said:
I know what you mean, it was more for a rough guide of how the total available performance would get affected from "new server A with no patch" to "new server A with patch". (Which is obviously quite a vague indicator but an indicator none the less.) I'm leaving the SQL guys to test their stuff as they have far more knowledge on SQL than me. I'm just coming at it from an Infrastructure point of view.
Click to expand...

If they/you haven't seen it, there is some guidance from MS about MS SQL Servers - https://support.microsoft.com/en-us/help/4073225/guidance-for-sql-server
 
Cromulent said:
OK so apparently my motherboard has an Overclocking profile thing in the BIOS where you can save your BIOS settings to a file on a USB drive. Am I right in thinking that to update the BIOS all I need to do is save the settings to USB, update the BIOS and then load the settings back again? I really hope the update process is easy.
Click to expand...

You need to check that. That is how it works, but a lot of those won't let you load settings in from a different version of the BIOS. ie it's fine for/loading saving different settings on the same BIOS, but won't help you if you update the BIOS and expect to load the old settings into the new BIOS.
 
You must log in or register to reply here.
Back
Top Bottom