Archive for November, 2008

Gary Smith & EE Times Must Read This Blog!

Tuesday, November 25th, 2008

Thanks to Sean Murphy who alerted me to the 2 part piece (here and here) in EE Times on Cloud computing and Software-as-a-Service for EDA.

Those of you who have been reading this blog have already heard me discuss this topic (several times)and many of you have already added your thoughts to the conversation.  It looks like Daya Nadamuni and Gary Smith must either read my blog or have independently realized how the EDA industry is moving.  Either way, given Gary Smith’s and EE Times’ street cred, this will likely help to accelerate any move in that direction.

There is a lot more to be said, so look for my next post where I’ll address the EE Times articles directly.

Meanwhile, Happy Thanksgiving to those of you in the US.

harry the ASIC guy

Facebook & Salesforce - What Does it Mean?

Tuesday, November 18th, 2008

Two weeks ago, at their annual Dreamforce Conference, Facebook and announced that they had jointly developed technology that will integrate within pages on the popular social networking site the enterprise apps from the Software-as-a-Service (SaaS) Customer Relationship Management (CRM) vendor. As an example of this integration, they demonstrated an app that can leverage the social aspects of Facebook to determine what “friends” on the service might be possible candidates for a job listing. As a result, recruiters can more easily reach a larger number of more qualified candidates and job seekers can be notified of potentially interesting job opportunities.  The app will automatically log where the referral came from and credit the friend with the referral.  Future integration with LinkedIn and MySpace should be forthcoming.

Besides improving the recruiting and job seeking process, there are 3 other aspects of this collaboration that are noteworthy and bring up possible applications in the EDA space.


First, as stated in Denis Pombriant’s CRM blog,

using Facebook’s widgets and up-to-date demographic data, companies can develop applications that leverage customer knowledge that enables them to better sponsor and understand communities of interest without the expensive and time-consuming effort of keeping a customer list current. By definition, a user of Facebook or other social site will keep his or her data current out of necessity, and this will move us a long way toward relieving the problems associated with aging lists and duplicate entries.”

Just as recruiters can find better potential job candidates, EDA vendors and design services companies can find better potential customers and clients. If you are a small EDA company that already uses their software, why not build an app on that links to Facebook or LinkedIn to find potential customers? Not for the purpose of spamming them (please!!!), but for the purpose of identifying those who might truly benefit from your products and services so you can contact them directly. Alternatively, if you’re not a customer, I’m sure that James Colgan would point out that there is also the professional user community at Xuropa that can serve a similar purpose. You could contact your prospect as follows:

“Hi Joe. My name is Harry Gries and I’m an independent ASIC Methodology Consultant. I noticed from your public profile that you are currently designing an extremely complex ASICs with some leading edge technologies and tools. Personally, I have over 20 years experience (over 14 years in the EDA industry) working with advanced technologies and methodologies and have helped several clients identify the right tools to use and put together working methodologies. If you feel you might have need for someone like me, please contact me through my profile.  Or, if you feel someone else might be interested, please pass on my contact info. Thanks for your time, Harry”.


The second interesting aspect is the increasingly acknowledged business application and utility of social media. Just today I reminisced with a co-worker about the early days of the Internet when web sites were blocked because employees would “waste time” surfing the net. He remarked that today, if we were to block access to the internet, we’d have legitimate outcries that employees could not get their job done. The internet has become indispensible, for googling a technical term, for accessing product information and users’ experiences, for keeping up-to-date on industry news and technical breakthroughs. I expect that social media sites, such as Facebook, will soon be acknowledged as just as indispensible as more traditional web sites today.


The third and final  interesting aspect is how the #4 provider in the CRM industry is “upsetting the applecart”. Continuing from Denis Pombriant’s CRM blog,

“…it was not any of the larger and older (and richer) software giants that made the announcement. Instead, it was a relative newcomer yet to celebrate its 10th birthday, and with revenues only a fraction of Oracle, SAP, or Microsoft, that made the news.

So why didn’t Oracle, SAP, or Microsoft, come to this conclusion and build a product? A good question. The answer rests less on technology — any of them could develop the technology — and more on temperament. Of the four companies, only has an on-demand or Software as a Service vision not clouded by the need to preserve a massive legacy code base and the considerable revenue stream it represents. In short, did this because it could and because it has a clear understanding of the future of computing.”

Three big industry leaders … the need to preserve a revenue stream … sound familiar?

harry the ASIC guy

Roles and Irresponsibilities

Wednesday, November 12th, 2008

Coffee Shop

This past Saturday I went to grab a cup of coffee at the local mom-n-pop coffee shop that I really like. There were no stirrers so I told the server that they were out.

“Yeah, we’ve been out for a few days. You can use a straw”.

“Why don’t you just go to the Smart and Final or the Von’s 30 feet away and grab a box”, I asked.

“I’m just the server, that’s not my job”.


On Sunday I was talking to Chuck, one of the other parents at the YMCA Adventure guides induction ceremony and boat regatta. Chuck works as an installer for AT&T and also does electrical jobs when he gets a chance.  He’s a hard worker, putting in six 12-14 hour days each week.

As we talked, I suggested that he could set out on his own, make more money and work less hours. Like Joe the Plumber.

“I don’t need the headaches. If something goes wrong on a job today, it’s not my problem. It’s my bosses problem. If I own a business, then it’s all my problem.  I’d rather someone just give me the job and I do what I’m told.”

My Client’s Large Company

I was trying to install Office Communicator the other day and there was something wrong with my account on the server. I called IT and the woman on the line tried to help me but could not figure out what was wrong.  So she closed my ticket, the one with her name on it, and opened a new one for a specialist in Office Communicator to look into it.

About a week later, someone else in IT called me up to help me with my issue.  He was able to figure out what was wrong, but lacked the permissions to make the fix.  So he closed out my ticket, the one with his name on it, and opened a new one for the person with permissions to fix the problem.

A few days later, I rebooted my laptop and Office Communicator was now working.  Later that day I got 3 emails from IT requesting me to fill out a short survey regarding the resolution of my issue.

A Person I Work With

The other day I urgently needed help to run an analysis on a chip I’m working on. So I asked one of the people on the team who knows how to do it quickly.

“That’s  not one of the things I’m responsible for.”


Am I the only person not afflicted by the “not my job” disease? Has this really become such an ingrained part of American and corporate life?

I’m sorry, I just don’t think that way. If I see a problem, then it’s my problem.  Maybe I’m anal or a perfectionist or neurotic and maybe I need to let go.  But I’m just not wired that way. And I don’t understand people who are.

harry the ASIC guy

A Scalpel or a Hatchet?

Thursday, November 6th, 2008

I spoke to a friend of mine at Cadence yesterday morning.

“I came in this morning and tried to log on and I couldn’t.  Turns out I must have mistyped my password, but I was worried there for a minute.”

That seems to sum up the mood at Cadence yesterday as the Turk made his way through the hallways of Cadence worldwide. “I think I’m going to go on vacation for the rest of the day”, my friend said later.

By now, I’m sure you’ve heard it from friends, or read it in EE Times or on John Blyler’s blog, that Cadence cut “at least” 625 jobs or 12% of it’s workforce.  John Blyler’s sources say the impact will be closer to 1000 jobs once contractors and others are added in.

The speculation has been going on for several weeks including an interesting exchange on the Yahoo message boards as to how deeply and in what manner Cadence would cut.  Would they take out a scalpel and trim the fat from organizations across the board?  Or would they focus on key areas with technological advantage and hatchet off areas that were non-competitive? Cadence CTO Ted Vucurevich mocked the idea that Cadence would sell off any businesses. Yet, the official press release says that Cadence “emphasized those market segments where Cadence enjoys a leadership position, such as mixed-signal design, advanced verification, and low-power design”, implying it was de-emphasizing other businesses.

So … do we have any idea if these cuts were made with a scalpel or with a hatchet? Here is what I’ve been able to pick up from others regarding what Cadence is doing to save expenses:

  • From the official 8-K report: “(e) Cadence has determined that no payment shall be made to Cadence’s named executive officers or the other participants in the Cadence Senior Executive Bonus Plan for performance in fiscal 2008.” I don’t think this eliminates the golden parachute, but at least it weighs a little less.
  • At least some middle management was let go, including first line managers and some long-time Cadence employees. It seems they are trying to cut the “deadwood” like any organization does now and again but from the management side first. And the impact to the rank and file seems to be less than it could have been. Still, having gone through similar times at Synopsys before, I know that a lot of these people are experienced EDA veterans and good people to have in the organization and I wish them well.
  • All, or a substantial part of their tech pubs was let go. (Makes sense since they can contract this out).
  • IT was hit, though not sure how deep. (Can always move their infrastructure to the cloud or outsource IT - now wouldn’t that be ironic).
  • Berkeley Labs was hit hard or perhaps even shut down (who needs research?)
  • Their DFT R&D group was impacted, perhaps a signal as to their plans for this product line (just a bunch of IBMers anyway).

As yet, there is no indication that entire products are going away, but it’s still early.  Cadence has yet to speak to the investment community since the Fister resignation and delaying their earnings release due to the discovery of an “accounting error”. Personally, I think it will be painful for a company that has endeavored to provide a “complete flow”, to sell off a product and create a hole in that flow.  If they do sell off products, it will be around the edges, not in the core EDA flow.

Take a look at Gary Smith’s analysis from the time that Cadence bid to take over Mentor (can you believe it’s been  just over 4 months since the offer). The tools where Cadence has low market share and are possible targets for the hatchet:

  • Design for Manufacturing
  • Physical Verification
  • Design For Test
  • Static Timing Analysis
  • Synthesis

What do you think? Can you see Cadence divesting any of these products?  Or do you think there are others that might go away? Which product would you definitely not let go?

Inquiring minds want to know….

harry the ASIC guy