A new approach to hardware design project management
Simon Butler, Methodics
EETimes (9/20/2011 3:44 PM EDT)
Introduction
Data management has always been a pesky ‘background’ problem for IC designers, who usually found ways to stitch together reasonably effective solutions. In the beginning there were files and directories, and developers would version data by maintaining copies with unique names. Later, this was replaced by RCS/CVS simplistic file versioning. Over the years, next-generation DM (data management) and CM (configuration management) tools such as Perforce, Subversion, ClearCase, and Git have emerged that improve performance and reliability. These tools added a layer of abstraction over the file versioning problem.
As organizations evolved and (with Moore’s Law) designs exploded in complexity and size, many were forced to resort to multi-site design. File-counts grew, file-sizes expanded, and multiple DM repositories and even multiple DM tools were often used on a single project. Today, many design organizations struggle to keep project-data organized properly and communicate change effectively. Finally, exacerbating the situation, companies suffered from poor or no permission management strategy, bad performance, inconsistent data management systems, and spiraling disk/network resource requirements. There was no single way to control, measure, or manage the situation.
To read the full article, click here
Related Semiconductor IP
- 12-bit, 400 MSPS SAR ADC - TSMC 12nm FFC
- 10-bit Pipeline ADC - Tower 180 nm
- NoC Verification IP
- Simulation VIP for Ethernet UEC
- Automotive Grade PLLs, Oscillators, SerDes PMAs, LVDS/CML IP
Related Articles
- Hardware Configuration Management and why it's different than Software Configuration Management
- Single core to multicore: Addressing the system design paradigm shift with project management and software instrumentation
- Why Hardware Root of Trust Needs Anti-Tampering Design
- System-on-chip (SoC) design is all about IP management
Latest Articles
- Analog Foundation Models
- Modeling and Optimizing Performance Bottlenecks for Neuromorphic Accelerators
- RISC-V Based TinyML Accelerator for Depthwise Separable Convolutions in Edge AI
- Exclude Smart in Functional Coverage
- A 0.32 mm² 100 Mb/s 223 mW ASIC in 22FDX for Joint Jammer Mitigation, Channel Estimation, and SIMO Data Detection