Tesla C1060 Memory Model

[firstborn]

I'm trying to found out a bit more info on the memory architecture of the latest Tesla cards.

The previous series 8 had say; 128 processors, grouped in blocks of 16, with each block having access to on chip high speed memory and all blocks having access to shared global memory. Hence to maximise computation speed up, data would need to be constrained to on chip shared memory as much as possible as read/writing to global mem had large overheads compared to on chip mem.

All I can find on the Tesla C1060 is that it has 240 processors with access to 4GB shared memory - is it as simple as that? 240 processors each with rapid access to the full 4GB??

Cheers

 

[firstborn]

Common, somebody must have an idea.

 

[Natima]

Form Factor 10.5" x 4.376", Dual Slot
# of Tesla GPUs 1
# of Streaming Processor Cores 240
Frequency of processor cores 1.3GHz
Single Precision floating point performance (peak) 933
Double Precision floating point performance (peak) 78
Floating Point Precision IEEE 754 single & double
Total Dedicated Memory 4GB GDDR3
Memory Speed 800MHz
Memory Interface 512-bit
Memory Bandwidth 102GB/sec
Max Power Consumption 200 W peak, 160 W typical
System Interface PCIe x16
Auxiliary Power Connectors 6-pin & 8-pin
Thermal Solution Active fan sink
Programming environment CUDA

Its just a GTX280 with much more memory.

 

[firstborn]

Cheers. Will hunt for GTX280 info. Basically need a hardware block diagram to confirm the memory set up. Below is the diagram for the G80 and really need similar for the G10 series.