EDP 2008: Multicore, Parallel Programming and Industry Outlook

• 
  EDP 2008 is an IEEE technical event that is run by the Design Automation Technical committee. The event is a 2 day workshop that invites papers from the semiconductor and systems companies, EDA vendors and the financial and analyst community that are part of the design automation ecosystem. The themes of the workshops tend to be topical and forward looking and focus on bringing to light technical issues that the industry needs to work towards resolving as a whole, including standards.
  
  This year’s EDP conference focused on the continuing evolution of multi-core systems and their associated design problems. Semiconductor vendor perspectives offered in the Intel and Sun keynotes reflected upon the difficulty of programming multicore platforms, the lack of software that could exploit the concurrency of the multicore design and the limiting nature of memory latency that holds down system speed. Speakers from ARM, MIPS, ARC, Tensilica, Sonics examined various aspects of design and implementation. The issue of how much software engineers would pay for a multicore toolset was a topic of debate. The consensus seemed to be that they would not pay very much. The problem then becomes a business issue more than a technical one.
  
  Underlying all this is the acceptance that multicore is real, it is here and it will become more pervasive going forward. The key issues facing the industry today, as several speakers pointed out, is how their solution or tool solves a problem or a set of problems associated with multicore design, integration, verification, test or other. At the end of the day, similar to any historical platform evolution of the model, the set of vendors that are able to provide standards based models for a multicore platform are likely to be the winners in the game. Make no mistake, the industry is going through a very important inflexion point. Whether you choose to call it ESL or multicore or many core or whatever, everyone must have a strategy for how they are going to participate in this new design ecosystem.
  
  While the overall macro climate might be shaping the current gloomy outlook on EDA, the VC outlook that EDA is an overinvested sector is debatable. Equally debatable is the view that it is all doom and gloom for EDA. With an expected outlook of 4% growth for 2008 and higher growth beyond, it does not appear that EDA is in the doldrums.
  
  Daya Nadamuni