Lisa Spelman | Google Next 2017
summary
Google Cloud runs the new Intel Xeon Skylake now, offered to cloud providers as an early, production-ready option that aims to cut TCO and boost workload-specific throughput. Key platform features include Skylake on Google Cloud, AVX-512 for HPC acceleration, Intel Resource Director tech for QoS and 72-core Xeon Phi for massive parallel jobs. FPGA (Altera) offload, Silicon Photonics, and Nervana’s neural silicon Lake Crest expand the portfolio; Intel targets >20% perf per generation and lists eight new product generations since the last x86 rival.
AI-written summary · about
chapters (8)
- 0:00Introduction to the cube
- 1:04Intel and Google partnership
- 3:26Xeon Skylake architecture
- 6:43Cloud service provider trends
- 10:12Hardware portfolio expansion
- 13:40AI solutions and strategy
- 16:48Competition and future growth
- 20:09Performance and closing
related talks
36mIntroduction to High Performance Computing on Google Cloud Platform
06mUsing FPGA's and Xeon processors to accelerate cloud workloads
33mReimagining the Modern Data Center
25mSustainability at the Core with Cloud Native Processors
01h 13mEngineering the Cloud for Tomorrow: Performance, Speed and Scale for AI Workloads
24mSilicon Development
42mHow Intel & AWS solve some of the world’s most urgent issues
56mKeynote on Building the Cloud for You