Inference Acceleration: Follow The Memory

By Geoff Tate - 07 Mar, 2019 - Comments: 0

Much has been written about the computational complexity of inference acceleration: very large matrix multiplies for fully-connected layers and huge numbers of 3x3 convolutions across megapixel images, both of which require many thousands of MACs (multiplier-accumulators) to achieve high throughput for models like ResNet-50 and YOLOv3. The other side of the coin is managing the movement of d... » read more

Use Inference Benchmarks Similar To Your Application

By Geoff Tate - 07 Feb, 2019 - Comments: 0

If an Inference IP supplier or Inference Accelerator Chip supplier offers a benchmark, it is probably ResNet-50. As a result, it might seem logical to use ResNet-50 to compare inference offerings. If you plan to use ResNet-50 it would be; but if your target application model is significantly different from Resnet-50 it could lead you to pick an inference offering that is not best for you. ... » read more

Real-Time Object Recognition At Low Cost/Power/Latency

By Geoff Tate - 01 Nov, 2018 - Comments: 0

Most neural network chips and IP talk about ResNet-50 benchmarks (image classification at 224x224 pixels). But we find that the number one neural network of interest for most customers is real-time object recognition, such as YOLOv3. It's not possible to do comparisons here because nobody shows a YOLOv3 benchmark for their inferencing. But it's very possible to improve on the inferencing per... » read more

Performance Benchmarking Embedded FPGAs

By Flex Logix - 03 Oct, 2018 - Comments: 0

When evaluating the performance of an embedded FPGA, one needs to evaluate the performance of each of the individual modules that make up an FPGA. The basic modules are: Reconfigurable logic building blocks (RBB-Logic), Fine-granularity logic containing LUTs, carry-forwarding adder chain, and flip-flops Reconfigurable DSP building blocks (RBB-DSP), Medium-granularity arith... » read more

System-Level Verification Tackles New Role

By Brian Bailey - 28 Apr, 2016 - Comments: 0

Wally Rhines, chairman and CEO of Mentor Graphics, gave the keynote at DVCon this year. He said that if you pull together a bunch of pre-verified IP blocks, it does not change the verification problem at the system level. That sounds like a problem. There are assumptions made that the IP blocks work to a reasonable degree, and that when performing system-level verification the focus is not a... » read more

Metrics For Measuring Performance And Power In IoT SoC Designs

By Emerson Hsiao - 08 Oct, 2015 - Comments: 0

The problem confronting chip designers developing IoT SoCs is the need for high compute performance and low power consumption. This is especially true for SoCs being developed for devices required to operate for years on a battery. One example is the new generation of electronic shelf label (ESL) with a requirement of five years. The ESL receives central server pricing updates along with a f... » read more

Newer posts →

Knowledge Centers
Entities, people and technologies explored

Partitioning In The Chiplet Era

Understanding how chiplets interact under different workloads is critical to ensuring signal integrity and optimal performance in heterogeneous designs.

by Ann Mutschler

3.5D: The Great Compromise

Pros and cons of a middle-ground chiplet assembly that combines 2.5D and 3D-IC.

by Ed Sperling

Legacy Process Nodes Going Strong

The critical, and growing, significance of mature node chips and processes.

by Bryon Moyer

AI’s Role In Chip Design Widens, Drawing In New Startups

Focus is on letting engineers do much more with the same or fewer resources — and less drudgery.

by Karen Heyman

Key Technologies To Extend EUV To 14 Angstroms

Alongside high-NA EUV will be better-performing photoresists, reduced roughness using passivation and etch, and lateral etching to reduce tip-to-tip dimensions.

by Laura Peters

Memory Fundamentals For Engineers

eBook: Nearly everything you need to know about memory, including detailed explanations of the different types of memory; how and where these are used today; what's changing, which memories are successful and which ones might be in the future; and the limitations of each memory type.

by The SE Staff

What Comes After HBM For Chiplets

The standard for high-bandwidth memory limits design freedom at many levels, but that is required for interoperability. What freedoms can be taken from other functions to make chiplets possible?

by Brian Bailey

Why Small Fab And Assembly Houses Are Thriving

Megafabs churning out the most advanced processors are not the only game in town.

by Bryon Moyer

tag: benchmarks

Inference Acceleration: Follow The Memory

Use Inference Benchmarks Similar To Your Application

Real-Time Object Recognition At Low Cost/Power/Latency

Performance Benchmarking Embedded FPGAs

System-Level Verification Tackles New Role

Metrics For Measuring Performance And Power In IoT SoC Designs

Trending Articles

Hybrid Bonding Makes Strides Toward Manufacturability

Americas Chip Funding Energizes Industry

Advanced Packaging Driving New Collaboration Across Supply Chain

Chiplets Make Progress Using Interconnects As Glue

2D Semiconductors Make Progress, But So Does Silicon

Knowledge Centers
Entities, people and technologies explored