OMG, I've seen the light, I must head towards it.
Thats simply awesome GPU power
Solitaire must play really well
Intel Xeon E5-2665 spicy for $260
- Thread starter Jeggs101
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Thats simply awesome GPU power
Solitaire must play really well
Solitaire must play really well
.... and checkers too ....
Still looking for a GPU update to MS Decathlon
Seriously,
the system "needs" more I/O bandwidth and a Xeon CPU (for one specific reason).
One Titan connected to a x16 slot can transfer approx. 11 GB/sec over the PCI Express connector. If connected to a x8 slot, the speed goes down to 5,5 GB/sec.
With a 5,5 GB/sec transfer speed, one full transfer of the GDDR RAM which is 6 GB, takes approx 1 sec - an eternity when a lot of CUDA cores are waiting for new data.
Another "limitation" is the aggregated bandwidth. All 4 cards have an aggregated bandwidth of 22 GB/sec on the X79 platform. While this looks like only 50-60% of the total available memory bandwidth the LGA2011 socket has, the i7-3930K CPU misses a critical capability of the Xeons. It doesn't have the Data Direct functionality.
Basically, all I/O operations writing to main memory must preserve memory consistency, which is done via writing through the cache. Writing 1 GB of I/O data through a 10-20 MB cache basically invalidates all cache data the CPU thought might be useful for future usage. Bad luck. It has to get its data from main memory again - which is significantly slower than reading from the cache. The performance of the i7 tanks as it doesnt get data fast enough. The SB Xeon's circumvent this bottleneck by splitting the cache. Only a part of the cache will be used for this fast I/O into main memory and the remaing part of the cache can still serve the CPU's desire for data.
Andy
Last edited:
What's holding you back then ?
The i7-3930K is a great CPU, but the difference between the XEON version of SB vs. the desktop comes to effect when high I/O rates are in there. I've stripped my dual E5-2687W workstation to a single socket with 64 GB to compare it to the i7/X79/64GB system. When 16 fast SSDs like the 256GB Samsung 840 Pro are pumping data into the system, the overall performance advantage of the Xeon over the i7 is between 50 and 60% - higher than the 2 additional cores would impact the result. As long as the the CPU memory hierarchy isn't "disturbed" the performance difference is much less. The 8 Xeon cores have to share the same memory bandwidth like the 6 cores of the i7, so it doesn't scale linearily with big data sets in memory. If the computational density is high enough that most of the data sets come from cache, then it is almost a simple CPU counting game (as they have the same micro-architecture).
Cheers,
Andy
Which stepping?
And Andy - have a few things that I do that are single core bound where the higher clock speed helps. Can always remote desktop into dual CPU boxes.
Don't know what steppings my chips will be. But I will run it for a few hours with some SW with algorithmic error-detection. It doesn't prove that the chip is error free, at least it can correctly compute numbers ...
Andy