Data Centers Hit the Accelerator
Reinvent the server, says Microsoft exec
Rick Merritt, EETimes
4/15/2016 01:00 PM EDT
SAN JOSE, Calif.—Kushagra Vaid wants a new kind of computer server, one more suitable for large-scale data centers and machine learning. The general manager of engineering of Microsoft’s cloud computing division is not alone.
Technical leaders from all the hyperscale data centers, processor vendors and many others are seeking the same Holy Grail—an architecture tuned for so-called deep neural networks. The emerging algorithms have been gaining traction in a wide variety of pattern-matching jobs from serving up Facebook news feeds to piloting self-driving cars and beating the human champion of Go.
To read the full article, click here
Related Semiconductor IP
- DC-DC Split-Pi Boost-Buck Converter
- Deep learning accelerator
- MIL-STD-1553 Controller IP
- UFS 5.x Device IP
- UCIe 3.x Controller IP
Related News
- IntelliProp First to Market with Memory Fabric Based on CXL; Driving Most Disruptive Technology to Hit Data Centers in Decades
- NeuReality and Veriest Achieve Great Engineering Feat to Advance AI Chips for World's Largest Data Centers
- VeriSilicon unveils the new VC9800 IP for next generation data centers
- Europe Leaps Ahead in Global AI Arms Race, Joining $20 Million Investment in NeuReality to Advance Affordable, Carbon-Neutral AI Data Centers
Latest News
- Tenstorrent Sets New Performance Records, Launches TT- Ascalon S, and Expands Across Japan
- Chips&Media Signs APV codec IP Licensing Deal with North American Big Tech, Establishing the ‘Second Front’ Against Apple’s ProRes
- Chipsolve Technologies Appoints Balaji Kanigicherla as Chairman of the Board
- OXMIQ Raises $35 Million to Scale OxCore™ Architecture
- SOC-E successfully showcases Time-Sensitive Networking (TSN) application for NGWS/FCAS Remote Carrier (Pillar 3)