Machine learning has begun to have a huge impact on the EDA Industry. Numerous organizations are heading forward with homegrown machine-learning algorithms to make specifications better.
Agnisys is making headway with home-grown machine-learning algorithms — and applying them to IDS NextGen multi-platform product which helps users create SoC specifications at an enterprise level.
It handles individual IP to sub-system to SoC level and is compatible with Word, Excel, IP-XACT, RALF, CSV, and SystemRDL. IDS NextGen generates design and verification codes for not just registers but sequences in one integrated environment. It reduces the verification time by generating the entire UVM SV and SystemC output Sequences.
The NextGen product attempts to “understand” the specification using Machine Learning technology and guides the user about issues with the specification. It helps create a standardized specification. Capturing issues in the specification is the extreme form of “Shift-Left” that the industry has been seeing. Agnisys motto is to stop issues from germinating in the first place so that less time is spent on the debug – which is often very costly.
Once the specification is entered, the user can create custom outputs using a template engine. IDS now supports all current prevailing input and output formats.
The NextGen product also supports special safety and reliability requirements for Automotive and IoT sectors.
Discover what’s new – IDS Next Gen
Comprehensive SoC/IP Specification and Code Generation Tool
-
Behavior Driven, Spec Centric Development
-
Single multi-platform UI product which helps users to create SoC specifications at an enterprise level
-
Promotes reuse and collaboration
-
Inbuilt DRC checks improve design quality
-
Inbuilt machine learning algorithms for guidance
-
Handles individual IP, sub-system to SoC level
-
Compatible with Word, Excel, IP-XACT, RALF, CSV, and System RDL.
-
Generates design and verification code for not just registers but sequences
-
Reduces the verification time by generating UVM SV, SystemC
Hence there are multiple opportunities and untapped resources in the EDA industry concerning machine learning. By turning mountains of raw data into valuable design insight, machine learning gives EDA a desperately needed shot in the arm.
