tag: NTHU

AI Architectures Must Change

By Brian Bailey - 13 Aug, 2018 - Comments: 4

Using existing architectures for solving machine learning and artificial intelligence problems is becoming impractical. The total energy consumed by AI is rising significantly, and CPUs and GPUs increasingly are looking like the wrong tools for the job. Several roundtables have concluded the best opportunity for significant change happens when there is no legacy IP. Most designs have evolved... » read more

Trending Articles

Chip Industry Week In Review

Applied Materials' $4B facility at risk; AI chip bonanza; Japan, U.S. alliance; Rapidus' U.S. subsidiary; EDA revenue up; TSMC $6.6B funding; Infineon, Amkor team up; S. Korea's big bet; Apple's processor overhaul; high-NA EUVL defect system; chiplet HW security module; new Spectre attack.

by The SE Staff

Chip Industry Week In Review

China restricts U.S. processors; Samsung's CXL, DRAM module and roadmap; ASML's $2.7B retention bonus; U.S., Mexico partner up; Keysight's $1.5B offer to Spirent; Lam's new pulsed laser deposition; new OSAT center.

by The SE Staff

Architecting Chips For High-Performance Computing

Data center IC designs are evolving, based on workloads, but making the tradeoffs for those workloads is not always straightforward.

by Ann Mutschler

What Works Best For Chiplets

Not all chiplets are interchangeable, and options will be limited.

by Anne Meixner

Chip Industry Week In Review

IC sales up; Taiwan's quake; Japan's big bets; Intel Foundry loss; edgeAI startup funding; smartphone market's recovery; robots; data center security; NoCs in 3D.

by The SE Staff

Knowledge Centers
Entities, people and technologies explored

Related Articles

The Rising Price Of Power In Chips

More data requires faster processing, which leads to a whole bunch of problems — not all of which are obvious or even solvable.

by Ed Sperling

The Future Of Memory

From attempts to resolve thermal and power issues to the roles of CXL and UCIe, the future holds a number of opportunities for memory.

by Karen Heyman

SRAM Scaling Issues, And What Comes Next

While it will remain a workhorse memory, using SRAM at advanced nodes requires new approaches.

by Karen Heyman

Verifying Hardware Security With RTL Simulation

Measuring how injected faults propagate through a design and how long they remain in the system.

by Taruna Reddy

Backside Power Delivery Adds New Thermal Concerns

Lack of shielding, routing issues, and new mechanical stresses could have broad impact on standard cell design.

by Karen Heyman

Architecting Chips For High-Performance Computing

Data center IC designs are evolving, based on workloads, but making the tradeoffs for those workloads is not always straightforward.

by Ann Mutschler

Glitch Power Issues Grow At Advanced Nodes

Problem is particularly acute in AI accelerators, and fixes require some complex tradeoffs.

by Ann Mutschler

Rethinking Memory

Von Neumann architecture is here to stay, but AI novel architectures and 3D structures create a need for new testing tools.

by Karen Heyman

About

Navigation

Connect With Us

Facebook
Twitter @semiEngineering
LinkedIn
YouTube

Copyright ©2013-2024 SMG | Terms of Service | Privacy Policy