Software Costs Skyrocketing for SOCs

By Dean ~ August 24th, 2011. Filed under: Systems Engr..

Semico Research Corp. recently released a report asserting that software design costs rose by 42.5% for SOCs implemented at the 28nm node.  The report goes on to forecast that software design costs will continue to rise at a CAGR of 138.9% through the 14nm node.  Wow, that’s going to hurt!  Thankfully, the EDA vendors are trying to step up their offerings, as described in the Steve Leibson‘s post on EDA360 Insider (h/t to Steve for highlighting this report).

Systems Engineering Design Flow

Systems Engineering Design Flow

As one would expect, complex embedded systems are also experiencing skyrocketing costs for software development.  This is partly due to the increasing capabilities of the underlying platforms, and partly due to the dramatically higher demand for functions to be implemented in software to support flexibility and relative ease of upgrading.

Fortunately, we also have a solution in the embedded system world.  Model Driven Systems Engineering (MDSE) is an ideal methodology for taming the software “beast.”

MDSE allows teams to make optimal design decisions regarding partitioning between hardware and software, interfaces and architectures.  Practitioners can manage the complexity of the entire system to ensure that the software component stays under control.  Properly crafted system models, combined with the discipline of MDSE, also support system implementation by:

  • Providing an infrastructure for Performance Engineering
  • Establishing a framework to allow for implementation and evaluation of software earlier in the cycle
  • Acting as a powerful tool to diagnose unanticipated issues
  • Serving as a test bench for “implementation-in-the-loop” simulation
  • Avoiding integration issues by doing trial integrations early & often within the system model.

As Paul so eloquently stated in the original post:

“The system model becomes not just an artifact but a powerful tool that dramatically improves communication, collaboration and execution throughout the product cycle.  Model Driven Systems Engineering is the key to deploying complex systems on spec, on time and on budget.”

Comments are closed.