How manycore will reshape EDA
Abha Maheshwari, Cadence Design Systems
3/7/2011 12:50 PM EST
It is difficult to imagine a world without the smartphones, tablets, e-readers and game consoles that are pervasive in our lives today. Behind such innovations are embedded systems. To provide the automation and the speed to build them, EDA companies are taking advantage of the parallelism offered by multiple cores. All claim some multithreading and multicore capabilities and are porting more code to multithreaded applications.
This is certainly no easy task. EDA code is complicated, and EDA developers must work within Amdahl’s Law, which states that the gains you get from parallelizing code are sharply limited by any section that cannot be parallelized. Thus, if 90 percent of the code is parallelized and 10 percent is not, the maximum gain expected is 10x.
To read the full article, click here
Related Semiconductor IP
- SHA-256 Secure Hash Algorithm IP Core
- EdDSA Curve25519 signature generation engine
- DeWarp IP
- 6-bit, 12 GSPS Flash ADC - GlobalFoundries 22nm
- LunaNet AFS LDPC Encoder and Decoder IP Core
Related Articles
- SoC realization: Finally the "Killer App" that will allow EDA to grow again?
- EDA in the Cloud Will be Key to Rapid Innovative SoC Design
- Software will be main SoC worry
- 2002 will bring more chip consolidation after worst year ever, says Dataquest
Latest Articles
- TensorPool: A 3D-Stacked 8.4TFLOPS/4.3W Many-Core Domain-Specific Processor for AI-Native Radio Access Networks
- Assertain: Automated Security Assertion Generation Using Large Language Models
- VolTune: A Fine-Grained Runtime Voltage Control Architecture for FPGA Systems
- A Lightweight High-Throughput Collective-Capable NoC for Large-Scale ML Accelerators
- Quantifying Uncertainty in FMEDA Safety Metrics: An Error Propagation Approach for Enhanced ASIC Verification