GOYA INFERENCE PERFORMANCE

BERT – Base SQuAD Performance

Goya performance on BERT-Base, a popular natural language processing model, is based on the following configurations:
Hardware: 1x GOYA HL-100; CPU Xeon Gold [email protected] GHz.
Software: Ubuntu v-18.04; SynapseAI 0.11.0-477

 


Performance on ResNet50

Goya delivers super fast throughput and ultra low latency on ResNet50, a popular image recognition model. These performance metrics are based on:
Hardware: Goya HL-100 PCIe card; host CPU Xeon Gold [email protected] GHz; and
Software: Ubuntu v-18.04, SynapseAI v-0.11-447.


Inference performance: All AI modelsInference published MLPerf results

GOYA INFERENCE PRODUCTS

GOYA PCIe Card

GOYA HL-100 PCIe CARD:

• PCIe: Gen4 x 16 lanes
• Form Factor: Dual-slot
• Memory: 16GB with ECC

GOYA HL-100 PCIe Card Datasheet
GOYA 8 card server

GOYA 8-CARD SERVER:

HL-100 built for industry standard 8-card servers

Download GOYA whitepaper

HABANA SynapseAI® SOFTWARE Suite for Inference

SynapseAIHabana’s software platform provides efficient and flexible development and deployment of a wide array of inference workloads and ease of customization to suit user-specific solution requirements. Designed for flexibility, Habana’s inference software leverages foundational and flexible blocks: Habana’s programmable Tensor Processor Core (TPC) SDK,  optimizer, graphic compiler and runtime software, a rich and extensible kernel library and data center deployment and management tools.

TPC KERNEL LIBRARY & SDK

  • Programmable TPC with SDK
  • Extensive TPC kernel libraries support user customization
  • TPC tools (simulator, compiler, debugger) for custom kernel development

TCP Habana

OPTIMIZER, COMPILER & RUNTIME

  • Seamlessly integrates with existing frameworks
  • Can be interfaced with C or Python API
  • Supports development flow from graph to optimized recipe
  • Enables quantization, pipelining, layer fusing

SynapseAI Habana

PERFORMANCE MANAGEMENT & DEPLOYMENT TOOLS

  • Deployment validation toolset: test and monitor functionality and performance
  • Libraries and tools for run-time detection and reporting
  • Run-time and orchestration plug-ins

Software Tools icon

HABANA INFERENCE SOFTWARE PLATFORM
SUPPORTS CUSTOMIZATION AT EVERY LAYER