editor's blog
Subscribe Now

Go Wide

Last week Cadence announced a new wide-I/O memory controller IP block, ostensibly the first of its kind. This actually represents a risk start based on a JEDEC standard that’s not yet complete.

The idea behind the wide-I/O movement is predicated on use in 3D ICs, where a memory chip will be stacked on a logic chip, with the connections being made by TSVs. Instead of requiring higher-drive I/Os that connect to chip pads and traverse PCB traces to get to a memory chip (or back from the memory chip), you stay entirely within the package. An array of TSVs mean that you can handle far more I/Os that if you have to go to package pins. And the drive requirements are reduced tremendously, reducing both the size (due to smaller transistors) and power of the resulting combination.

Of course, with more connections, you get much higher bandwidth: this is a 512-bit interface. That’s a lot more data available in one chunk than you can traditionally get.

Cadence’s controller block includes traffic shaping algorithms to increase throughput as well as features to address power, including traffic sensing (so that power can respond to traffic) and an option for dynamic voltage and frequency scaling (DVFS).

This would seem to come well ahead of the standard, which is projected (no promises!) to be available to non-members in September. But, in many such standardization cases, the technical details are approved first, and then the resulting standard goes through a higher-level board approval step that largely examines the process by which the standard was set to make sure that it was done properly. 

Clearly Cadence is betting that there will be no further technical changes. Or that, if there are, they can update the IP before any customer commits to final silicon.

Leave a Reply

featured blogs
Nov 12, 2024
The release of Matter 1.4 brings feature updates like long idle time, Matter-certified HRAP devices, improved ecosystem support, and new Matter device types....
Nov 13, 2024
Implementing the classic 'hand coming out of bowl' when you can see there's no one under the table is very tempting'¦...

featured video

Introducing FPGAi – Innovations Unlocked by AI-enabled FPGAs

Sponsored by Intel

Altera Innovators Day presentation by Ilya Ganusov showing the advantages of FPGAs for implementing AI-based Systems. See additional videos on AI and other Altera Innovators Day in Altera’s YouTube channel playlists.

Learn more about FPGAs for Artificial Intelligence here

featured paper

Quantized Neural Networks for FPGA Inference

Sponsored by Intel

Implementing a low precision network in FPGA hardware for efficient inferencing provides numerous advantages when it comes to meeting demanding specifications. The increased flexibility allows optimization of throughput, overall power consumption, resource usage, device size, TOPs/watt, and deterministic latency. These are important benefits where scaling and efficiency are inherent requirements of the application.

Click to read more

featured chalk talk

Industrial Internet of Things
Sponsored by Mouser Electronics and CUI Inc.
In this episode of Chalk Talk, Amelia Dalton and Bruce Rose from CUI Inc explore power supply design concerns associated with IIoT applications. They investigate the roles that thermal conduction and convection play in these power supplies and the benefits that CUI Inc. power supplies bring to these kinds of designs.
Aug 16, 2024
50,900 views