Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 11 years ago.
Improve this question
lets say i have a dual core having a speed of 2.7. Does the 2.7 stands for the sum of the speed of each core, or the speed of each individual core?
It means that each CPU core runs 2.7 billion cycles per second. This has a lot less meaning than it used to, as the amount of "work" that is completed each cycle varies quite a bit (due to considerations like caching, pipelining, hyperthreading, memory access times, and so on).
If you want to know how fast a processor is, it is much more advisable to look at benchmarks related to the kind of tasks you are trying to accomplish with it than to look at the clock speed. Consider: a dual-core 2.4 GHz Core i5 is much, much faster than a 2.4 GHz Pentium D (also dual-core).
Related
Closed. This question is not about programming or software development. It is not currently accepting answers.
This question does not appear to be about a specific programming problem, a software algorithm, or software tools primarily used by programmers. If you believe the question would be on-topic on another Stack Exchange site, you can leave a comment to explain where the question may be able to be answered.
Closed 2 days ago.
Improve this question
I'm working on a virtual machine that has been given 12 GB of RAM and has a Quadro RTX 6000 with 24 GB of VRAM. I'm trying to do machine learning on this virtual machine.
My intuition right now suggests that it is not using the video card as efficiently as it could be with the limited amount of RAM, and would work better with more. Is this correct?
The following suggests it is so, but is not very clear on that.
OpenCL - what happens if GPU memory is larger than system RAM
In short, how much RAM should I expect to need for machine learning and computer vision for this video card typically?
It very much depends on the software you are using. In some cases, GPU software can use significantly more VRAM than RAM, when the model runs only on the GPU and there is no need to have a copy of it in RAM.
As an example, although CFD and not ML, the FluidX3D software uses between 3.2x and 5.4x more VRAM than RAM. Here in your case the 24GB VRAM capacity would still be the limiting factor.
If it's 1:1 RAM:VRAM allocation, then you're limited by the 12GB RAM. In the end, you have to test your software and check the allocation ratio with tools like top/htop and nvidia-smi.
Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 7 years ago.
Improve this question
If the speed of the CPU is doubled, would the system be able to handle twice as many processes? Assuming you ignore context switches that is.
No. CPU speed is rarely the bottleneck anymore. Also, doubling the clock speed would require changers in both your OS's scheduler and your compiler (both of which make assumptions about the speed of a CPU relative to the data buses).
It would make things better, but it's not a linear improvement.
Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 10 years ago.
Improve this question
I was looking at the Azure full calculator and did not see a slider for adjusting the memory. Do they allow customized memory size or am I stuck with these silly per-configurations? I want at the minimum 16GB with the ability to go much higher like 300GB+.
There are five sizes (extra small, small, medium, large, and extra large), and they're fixed. Extra large is a 16GB VM, but after overhead of the OS, you only get 14GB.
Other sizes may appear over time, but it's unlikely you'll see 300GB any time soon. Windows Azure uses commodity servers, and they don't have nearly that much RAM.
Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 10 years ago.
Improve this question
In a design I am currently working on, I need quad port ram. However implementing it in lookup tables is using a massive amount of area and I cant reach the needed performance with that setup. Since, my FPGA has hardware blocks for single and dual port ram, is there anyway I can combine them to make quad port memory?
You could consider double-clocking the block RAM, although this will have implications for timing, etc.
See e.g. http://www.xilinx.com/support/documentation/application_notes/xapp228.pdf.
If you only need quad read access, then you just need two dual-port block RAMs, both connected to the same write-enable and write data.
Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 10 years ago.
Improve this question
What factors determine the maximum amount of physical memory a system can have? I know the operating system and the hardware play a roll. For hardware does it come down to the number of lines in the address bus?
According to this website different versions of Windows 7 have differnet caps on the usable memory, but what I don't get is why they don't corolate with the 32 bit/64 bit version? For example why does Home Premium support up to 16GB and Professional up to 192GB when both are 64 bit?
Is the max the lower of the two: hardware and operating system? For example what would happen if you had a 32bit address bus and put in 8GB of ram and had Windows 7 64bit?
Any 32 bit systems will not recognise anything more than 4GB. You can put whatever you want in it, but you will only get 4GB usable (actually a little less).
Regarding Windows, it is merely a marketing restriction. If Home Premium would support 192GB who would buy Professional?
It like if your Toyota Yarris could do 300 km/h in 10 seconds why would you buy Ferrari? :-)