Accelerators and GPUs Flashcards
Why do CPUs need to balance multiple factors in their design?
CPUs must deliver acceptable performance for a wide range of applications while balancing functionality, performance, energy efficiency, and cost.
What is the purpose of an accelerator in computing?
An accelerator works alongside a CPU to provide increased performance for specific workloads, with different design tradeoffs.
Give an example of an early accelerator used in CPUs.
x87 floating point co-processors were used to accelerate floating point operations before floating point units were integrated into the main processor.
What are GPUs originally designed for?
GPUs were specialised for image generation, requiring many matrix-vector operations.
Why are GPUs highly parallel?
They contain a large number of floating point units and support a large number of processing threads.
How does GPU memory differ from CPU memory?
GPUs use a different type of memory that provides higher bandwidth than CPU memory.
What was a key hardware evolution in GPUs?
GPUs evolved from fixed-function rendering pipelines to programmable unified shaders and double-precision arithmetic.
Why are GPUs useful as accelerators in HPC?
They offer good floating point performance and high memory bandwidth, making them suitable for computationally expensive tasks.
Why must some operations still be handled by a CPU in a GPU-accelerated system?
The CPU is responsible for tasks such as running the operating system and handling input/output operations.
How are GPUs typically connected to CPUs?
Via the PCI Express (PCI-e) interface.
What is a drawback of PCI-e connectivity for GPUs?
It has relatively high latency, which can impact performance.
How can the performance impact of PCI-e latency be mitigated?
By minimising the transfer of data between the host CPU and the GPU.
What technology provides better GPU connectivity than PCI-e? What makes it better?
NVIDIA’s NVLink technology, which offers a better data rate.
Why is energy efficiency important in HPC system design?
Power consumption and cooling requirements significantly impact overall system performance and cost.
What is the Green500 list?
A companion to the Top500 list that ranks HPC systems by energy efficiency (GFlops/Watt).
What is the exascale era in computing?
The era of building exascale supercomputers, which perform at least one exaflop (10^18 floating point operations per second).
Name the first exascale supercomputer.
Frontier (AMD EPYC CPUs and AMD Radeon GPUs).
Why is hardware diversity increasing in HPC?
Companies beyond Intel and NVIDIA, such as AMD and ARM, are entering the market with competitive CPUs and GPUs.
Why is portability an important consideration in modern HPC?
The increasing variety of CPU and GPU architectures means software must be able to run across different hardware platforms.
What is the NVIDIA Pascal architecture?
A GPU architecture designed for HPC applications.
How many Streaming Multiprocessors (SMs) can a Pascal GPU have?
Up to 60 SMs, though some may be disabled due to manufacturing defects.
What is a Graphics Processing Cluster (GPC)?
A block of 10 Streaming Multiprocessors, functioning like an independent GPU within a Pascal GPU.
How much high-bandwidth memory does a Pascal GPU have?
16GB.
