What is speedup in throughput for each of these improvements

Assignment Help Electrical Engineering

Reference no: EM13218085

Assume a GPU architecture that contains 10 SIMD processors. Each SIMD instruction has a width of 32 and each SIMD processor contains 8 lanes for single-precision arithmetic and load/store instructions, meaning that each non- diverged SIMD instruction can produce 32 results every 4 cycles.Assume a kernel that has divergent branches that causes on average 80% of threads to be active. Assume that 70% of all SIMD instructions executed are single-precision arithmetic and 20% are load/store. Since not all memory latencies are covered, assume an average SIMD instruction issue rate of 0.85. Assume that the GPU has a clock speed of 1.5 GHz.

Questions :

(1) Compute the throughput, in GFLOP/sec, for this kernel on this GPU.

(2) Assume that you have the following choices:

(1) Increasing the number of single precision lanes to 16

(2) Increasing the number of SIMD processors to 15 (assume this change doesn't affect any other performance metrics and that the code scales to the additional processors)

(3) Adding a cache that will effectively reduce memory latency by 40%, which will increase

instruction issue rate to 0.95

What is speedup in throughput for each of these improvements?

Reference no: EM13218085

Questions Cloud

Involvement in the insurance process : What are the implications of simultaneous federal and state involvement in the insurance process?

What current in amperes will flow through the bulbs : A 100 W bulb takes 0.833A and a 200 W bulb takes 1.67 A from a 120V source. If the two bulb were connected across a 240 V source, what current in amperes will flow through the bulbs?

Describe the muslim presence in africa in 18th-19th century : Describe the Muslim presence in Africa in the 18th and 19th century. How did Muslims make their money? How did they spread their influence? What areas of Africa were under Muslim control?

Calculate the amout of mmf generated by a coil : calculate the amout of mmf generated by a coil of wire having 1300 turns and carrying milliamperes of current?

What is speedup in throughput for each of these improvements : Increasing the number of SIMD processors to 15 (assume this change doesn't affect any other performance metrics and that the code scales to the additional processors)

The treaty of guadalupe hidalgo : The Treaty of Guadalupe Hidalgo... George Washington's success as a general is most accurately explained by.Benjamin Franklin.

What type of cable is necessary for each connection : What type of cable is necessary for each connection: straight or crossover? You can assume that S1 and S2 do not have the ability to resolve crossovers (called Auto-MDIX).

Calculate the minimum sampling interval : calculate the minimum sampling interval (time period for counting the pulse train) to achieve a speed resolution of 5 rpm, and the number of bits that must be there in the binary counter if the pulse train is to be sampled with this interval for a..

User Account

All Pages