Tech News
Back to home
Technology

Benefits And Challenges Of Using Chiplets

Source

SemiEngineering

Published

TL;DR

AI Generated

Moving to chiplets allows for more features in a smaller space, potentially leading to increased processing power, simpler designs, and higher yields. However, integrating chiplets is complex and involves considerations like coherent versus non-coherent interfaces and the choice between heterogeneous and homogeneous chiplets. Ashley Stevens from Arteris discusses the challenges and impacts of using chiplets in designs.

Read Full Article

Similar Articles

SemiEngineering

Optimal Heterogeneous Memory Configs for AI Tasks Under Specified Performance Metrics (Stanford, UCSC)

Researchers from Stanford University and the University of California, Santa Cruz have introduced a paper on "Heterogeneous Memory Design Exploration for AI Accelerators with a Gain Cell Memory Compiler." This paper focuses on the importance of on-chip memory systems that combine different technologies for optimal performance in AI tasks. The use of Gain Cell RAM (GCRAM) is highlighted for its advantages in density, power efficiency, and tunable retention compared to traditional SRAM. The OpenGCRAM compiler developed supports both SRAM and GCRAM, enabling the systematic identification of the best heterogeneous memory configurations for AI applications based on specific performance metrics.

SemiEngineering•
Distributed Authentication Framework Leveraging Multi-Party Computation In A Scalable Tree-Based Architecture (Univ. of Central Florida, Louisiana State)

Distributed Authentication Framework Leveraging Multi-Party Computation In A Scalable Tree-Based Architecture (Univ. of Central Florida, Louisiana State)

Researchers from the University of Central Florida and Louisiana State University have introduced a new distributed authentication framework called AuthenTree, leveraging multi-party computation (MPC) in a scalable tree-based architecture. This framework aims to enhance security in chiplet-based heterogeneous systems by enabling secure chiplet validation without exposing raw signatures, distributing trust across multiple integrator chiplets. AuthenTree has shown minimal overhead, with an area as low as 0.48%, an overhead power under 0.5%, and an authentication latency below 1 microsecond, surpassing previous solutions. The framework is designed to address security threats in post-fabrication environments and establish a more efficient and scalable security solution for next-generation chiplet-based systems.

SemiEngineering•
China's 96-core x86 CPU taps chiplet design to rival AMD EPYC and Intel Xeon — 13 chiplets per processor provide up to 384 cores on a single motherboard, but no word on power consumption

China's 96-core x86 CPU taps chiplet design to rival AMD EPYC and Intel Xeon — 13 chiplets per processor provide up to 384 cores on a single motherboard, but no word on power consumption

Zhaoxin has unveiled its next-generation Kaisheng KH-50000 processors, featuring a chiplet design with up to 96 cores, rivaling AMD and Intel server chips. The KH-50000 supports 12 channels of DDR5-5200 RAM, 128 PCIe 5.0 lanes, and 16 PCIe 4.0 lanes. It also includes support for x86 instructions, virtualization, and China's encryption standards. The processor's power consumption and TDP remain undisclosed, but its design allows for efficient utilization of older process nodes. Despite potential challenges, Zhaoxin aims to make significant strides in the domestic market with its latest server chip.

Tom's Hardware•
SemiEngineering

Chiplet Design Considerations

Chiplets are being used to increase compute capacity and I/O bandwidth by splitting SoC functionality into smaller dies integrated into a single system in package (SIP). Designers must consider system partitioning, process node selection, and die-to-die connectivity when designing chiplets. Advanced packaging options like interposers and RDL interposers with silicon bridges present new challenges in mechanical form factors, signal integrity, and power integrity analysis. Designers also need to focus on security considerations like authentication, data encryption, and secure boot processes in multi-die designs. System-level simulation and collaboration with companies like Synopsys can help optimize chiplet subsystem integration for efficient time-to-market delivery.

SemiEngineering•

We use cookies

We use cookies to ensure you get the best experience on our website. For more information on how we use cookies, please see our cookie policy.