Elma JetKit-3010, Руководство пользователя, Страница: 12 / 44

USER MANUAL JetKit-3010 ELMA.COM
www.elma.com EMBEDDED BOARDS
Features Specifications
Inte
rfa
ces
Extension
Interface 1x High-Speed Mezzanine Header:
- 1x PCIe x4
- 1x PCIe x2
- 1x USB2.0
- 4x CSI Camera Interfaces
- Utility signals
- Power
Bo
a
rd
Environmental Operating Temperature: -20°C to +50°C (@40W power profile)
Storage Temperature: -20 to +85°C
Management Integrated BMC controller handles reset management, power and
voltage monitoring
2.4 Features 1
2.4.1 Volta GPU
The same Volta GPU architecture that powers NVIDIA ® high-performance computing (HPC)
products was adapted for use in Jetson AGX Xavier™ series modules. The Volta architecture
features a new Streaming Multiprocessor (SM) optimized for deep learning. The new Volta SM is far
more energy efficient than the previous generations enabling major performance boosts in the
same power envelope. The Volta SM includes:
- New programmable Tensor Cores purpose-built for INT8/FP16/FP32 deep learning tensor
operations; IMMA and HMMA instructions accelerate integer and mixed-precision matrix-
multiply-and-accumulate operations.
- Enhanced L1 data cache for higher performance and lower latency.
- Streamlined instruction set for simpler decoding and reduced instruction latencies.
- Higher clocks and higher power efficiency.
The Volta architecture also incorporates a new generation of its memory subsystem and enhanced
unified memory and address translation services that increases memory bandwidth and improves
utilization for greater efficiency.
The Graphics Processing Cluster (GPC) is a dedicated hardware block for compute, rasterization,
shading, and texturing; most of the GPU’s core graphics functions are performed inside the GPC. It
is comprised of Texture Processing Clusters (TPC), with each TPC containing two SM units, and a
Raster Engine. The SM unit creates, manages, schedules and executes instructions from many
threads in parallel. Raster operators (ROPs) continue to be aligned with L2 cache slices and
memory controllers. The SM geometry and pixel processing performance make it highly suitable for
rendering advanced user interfaces; the efficiency of the Volta GPU enables this performance on
devices with power-limited environments.
Each SM is partitioned into four separate processing blocks (referred to as SMPs), each SMP
contains its own instruction buffer, scheduler, CUDA cores and Tensor cores. Inside each SMP,
CUDA cores perform pixel/vertex/geometry shading and physics/compute calculations, and each
Tensor core provides a 4x4x4 matrix processing array to perform mixed-precision fused multiply-
add (FMA) mathematical operations. Texture units perform texture filtering and load/store units
fetch and save data to memory. Special Function Units (SFUs) handle transcendental and graphics
interpolation instructions. Finally, the PolyMorph Engine handles vertex fetch, tessellation, viewport
transform, attribute setup, and stream output.
Table 2 - GPU Operation
GPU Configuration Performance
(peak) Max. Operating
Frequency per Core
# of TPC CUDA Cores Tensor Cores
4 512 64 10 TFLOPS | 30 TOPS 1.21GHz
1
Feature description taken from NVIDIA
®
Jetson AGX Xavier
™
Series System-on-Module Data Sheet.