Monthly Archives: January 2012

Welcome to Herb Sutter’s Jungle

In an effort to keep posting something here until I’m in the right place mentally to write about things that probably interest you, my dear friends, family, and online diabetes peeps, here’s another computing performance excerpt and link. (Working on this stuff is the 9-5 part of your favorite international playboy’s life.)


A half-decade after Herb Sutter wrote that the “free lunch” of Moore’s Law is over, he’s back with his prophet’s wisdom about where we’re going in his January Dr. Dobbs article, “Welcome to the Jungle”. I’ll give you a moment to decide whether to get the Guns N’ Roses song out of your head or use it as a backdrop for this juicy quotation:

If hardware designers merely use Moore’s Law to deliver more big fat cores, on-device hardware parallelism will stay in double digits for the next decade, which is very roughly when Moore’s Law is due to sputter, give or take about a half decade. If hardware follows Niagara’s and MIC’s lead to go back to simpler cores, we’ll see a one-time jump and then stay in triple digits. If we all learn to leverage GPUs, we already have 1,500-way parallelism in modern graphics cards (I’ll say “cores” for convenience, though that word means something a little different on GPUs) and likely reach five digits in the decade timeframe.

But all of that is eclipsed by the scalability of the cloud, whose growth line is already steeper than Moore’s Law because we’re better at quickly deploying and using cost-effective networked machines than we’ve been at quickly jam-packing and harnessing cost-effective transistors. It’s hard to get data on the current largest cloud deployments because many projects are private, but the largest documented public cloud apps (which don’t use GPUs) are already harnessing over 30,000 cores for a single computation. I wouldn’t be surprised if some projects are exceeding 100,000 cores today. And that’s general-purpose cores; if you add GPU-capable nodes to the mix, add two more zeroes.

The big takeaway for software engineers like me is that we’d best be learning how to develop solutions using the emerging APIs so that we can harness all of those extra orders of magnitude of scalability. That involves figuring out how to . . .

  • Deal with the processor axis’ lower section [of Sutter's chart] by supporting compute cores with different performance (big/fast, slow/small).
  • Deal with the processor axis’ upper section by supporting language subsets, to allow for cores with different capabilities including that not all fully support mainstream language features.
  • Deal with the memory axis for computation, by providing distributed algorithms that can scale not just locally but also across a compute cloud.
  • Deal with the memory axis for data, by providing distributed data containers, which can be spread across many nodes.
  • Enable a unified programming model that can handle the entire [memory/locality/processor] chart with the same source code.

Perhaps our most difficult mental adjustment, however, will be to learn to think of the cloud as part of the mainstream machine — to view all these local and non-local cores as being equally part of the target machine that executes our application, where the network is just another bus that connects us to more cores. That is, in a few years we will write code for mainstream machines assuming that they have million-way parallelism, of which only thousand-way parallelism is guaranteed to always be available (when out of WiFi range). . . .

If you haven’t done so already, now is the time to take a hard look at the design of your applications, determine what existing features — or better still, what potential and currently unimaginable demanding new features — are CPU-sensitive now or are likely to become so soon, and identify how those places could benefit from local and distributed parallelism. Now is also the time for you and your team to grok the requirements, pitfalls, styles, and idioms of hetero-parallel (e.g., GPGPU) and cloud programming (e.g., Amazon Web Services, Microsoft Azure, Google App Engine).


p.s. — I can’t believe that it’s been almost four years since I took a course with Herb out in Washington. That was some hard-core learnin’.

Posted in Computing, Fodder for Techno-weenies, Software Engineering | Leave a comment

We Need a New Mindset

Guy Steele drops a truth bomb.

(From How to Think about Parallel Programming: Not!)

Posted in Computing, Fodder for Techno-weenies, Software Engineering | Leave a comment

Thinking Differently about Software Optimization

Yesterday morning while eating my “Free Wednesday Breakfast” chocolate croissant and fresh fruit with yoghurt, I watched an interview with John Nolan entitled “The State of Hardware Acceleration with GPUs/FPGAs, Parallel Algorithm Design.” In the spirit of giving back, I’m posting a few notes.

  • When optimizing code for GPU, FPGA, or CPU, definitely focus on pipelining and overall throughput, not just local optimizations.
  • There’s a trade-off between “faster” and “sooner.” It’s not always worth saving a few seconds (or even a few minutes) if the kernels take hours or days to compile. (Then again, sometimes it is.)
  • Try to reduce dependence on the language/compiler “stack” that removes inefficiencies. The optimizer does good work, but you can do things to help it. Think about the hardware or architecture format. It’s not a sin to reduce the amount of abstraction in the service of performance. Pay attention to things that affect processor pipelining and cache movement.
  • BTW, some languages and technologies exist to provide higher level programming that’s close to the hardware, but they’re proprietary, secret, or still in R&D.
  • Use algorithmic optimization techniques. Step back and find the shortest-time computation.
  • Avoid using if statements. The goto construct is considered harmful, but if is basically the same thing. Instead think about state machines and polymorphism. There’s no branch-prediction penalty to pay, since the system “just is” in the state it’s supposed to be in. The logic is clearer, because there are no switches, making it easier to test, too.
  • Don’t always assume that floating-point values are necessary. Integers can often be creatively used and are far faster for math than double-precision numbers.
  • Of course, there’s a compromise between speedy/efficient and readable/maintainable.
  • Aim to structure programs as “symbolic intent.” Mathematical descriptions are bad ways of expressing programs. Think about functional programming models instead of procedural.

If you want to know more, you should definitely watch the half-hour interview. And if your reaction was more along the lines of “Yes, yes; that’s all true, and it’s how I design my image processing code,” then I definitely hope you’ll consider applying for the GPU/multicore engineering position we have open.

Posted in Computing, Fodder for Techno-weenies, From the Yellow Notepad, Software Engineering | Leave a comment

Now Hiring Image Processing Software Engineers

My group at work—the Image Processing and Geospatial Computing Group at MathWorks—is hiring a couple of software engineers. One of them could be you.

We need someone with GPU and multicore programming skills. We’re looking for experience with CUDA, OpenCL, OpenMP, Intel’s Threading Building Blocks, or similar technologies. If you’re into making algorithms run wicked fast, you should definitely apply.

The other position focuses on image processing and code generation. If you like implementing image processing algorithms and converting MATLAB code to C code, then this is the job for you.

I’ve been at The MathWorks for almost fourteen years now, and it’s a really great company with an excellent corporate culture, competitive compensation, fantastic benefits, and lots of perks. Because everyone uses MATLAB and because we’ve made some very sensible business decisions over the last 28 years, it’s a very stable company to work for. (Did I mention that we’re putting up our fourth building in our Natick campus? And I think I also mentioned that the entire worldwide staff went on a cruise a few years ago.)

If image processing isn’t your thing, we have dozens of other positions open. Everything from web development to legal department work. Human resources to customer service. Technical writing to application engineering and consulting. Marketing to program management. QE, sales, usability, and more software development positions than you can shake a stick at.

Come, help us accelerate the pace of engineering and science worldwide. And if you do apply, tell them I sent you.

Posted in General, Software Engineering | Leave a comment

Age of Majority

Eighteen years ago today, Lisa and I had our first “official” date. We watched Bill Clinton give hist first “State of the Union” address to Congress. Aww . . .

Now, our love has finally reached the age of majority. It can vote, form contracts, go to the casino, and buy lottery tickets and cigarettes legally. It no longer has the luxury of being sent to juvie for minor infractions. In New Hampshire it can ride a motorcycle without a helmet. In Puerto Rico and the U.S. Virgin Islands, it can buy its own booze.

It’s a brave new world for our relationship.

Posted in General | 1 Comment

To Paula Deen

Hey Paula!

It’s true I’ve never watched any of your shows or tried any of your recipes, but I saw your picture on a magazine cover or two at the supermarket. Now that you’ve announced that you have type-2 diabetes, I feel like we’re definitely on a first-name basis.

Type-1, type-2, LADA, gestational—no matter the kind, diabetes sucks. I’m sad to hear that you joined our club. It’s a bummer, but there’s a really big supportive group of people online and in real life who are here for you.

I know there’s a brouhaha brewing about the how/why/when of your announcement, but I frankly don’t care. . . .

Except I will say this: You owe us.

You have a high profile because of your pre-diabetes life. And you have partnered with Novo Nordisk to promote pharmaceuticals, putting you squarely in the diabetes community. You best be using your influence to help people with diabetes. Here are some ways:

  • Promote understanding of the differences between type-2 (which you have) and type-1 (which I have), since so many people in the US think there’s just one kind.
  • Encourage healthy lifestyle choices for everyone, whether touched by diabetes or not. T-2 is more than diet and exercise, but we have to be honest about their role. There’s no guarantee one way or the other, but every little thing we do (within reason) makes a difference.
  • Help people with diabetes in your audience understand that they are more than their disease, that there will be better days and worse days, and that they can do this.
  • Work with CDEs (at Novo and elsewhere) to develop a message of empowerment that people with diabetes can use to improve their own self-management as they make choices and work with their family and healthcare providers.
  • Stress that there’s more to diabetes than Novo’s drug-of-the-day. Be holistic.

Remember that a bunch of people are watching you. The diabetes online community is watching, too. We’re nice people, but we look out for our own. Based on what I’ve seen in the past, we will cut you like a piece of pecan pie if we think you’re using diabetes for your own benefit and aren’t giving back.

Diabetically yours!

Posted in Diabetes | 4 Comments

The Post Where I Talk Myself out of Seasonal Affective Disorder

Winter does funny things to me.

Starting around Christmas time I start to feel a bit overwhelmed. New prezzies (usually) means new books to add to my reading list. Extra time off work means more opportunities to clean up the detritus of the previous year (or longer). That’s a mixed blessing: freeing up space in my brain to concentrate on the right things without actually getting to spend the time doing those things. I’m being much more ruthless about just chucking stuff this year than in the past, and I think I’ll be done soon.

Almost being done is very good, because I have goals. (I don’t go in for New Year’s resolutions. Anything worth doing is worth starting at any point in the year. Why wait for a particular date to have a clean slate?) I tend to keep my goals to myself, but I’m willing to say that one of them involves trying to pimp-slap my out-of-control bookshelf by reading a certain number of pages each week. I figure that even an incredibly slow reader such as myself should be able to average 15 pages/day.

This goal-thinking was (is?) getting me a little down this year. So much of what I want to do in 2012 involves feats of athletic prowess, but my feet were threatening to get in the way of those feats. Lisa, the awesome exercise psychologist of my dreams, is (slowly) helping me see that I am more than my goals and accomplishments, but I still missed running because I really like it.

The week before my injury, I had a wonderful 12+ mile run that took me to the end of one branch of our local rail-trail and then past it into the exurban farmland and acreages of the neighboring towns before picking up the start of the other branch of the trail and following it home. I am eager to get back to that.

For sure, I was was also stressing that not doing these long training runs might leave me ill-prepared for the Around the Bay 30K in late March . . . or possibly incapable of running it at all. Eventually I told myself that I had to stop worrying about whether or not I would be able to do ATB—or the NYC Tri in July or the half-Ironman in August—and just concentrate on getting well. I could still ride my shiny new bike in the basement, there’s always plenty of swimming to do, and on the last day of work in 2011 I got a personalized weight-training program, which I started last week.

Sometimes I need to be reminded to look at the “big picture.”

By the middle of last week my foot didn’t really hurt very much, although I noticed twinges now and again, especially when I moved my foot in particular ways. It kinda sounded like plantar fasciitis, and it kinda didn’t. Everyone I talked to about it had horror stories about how PF messed up a fellow runner for months or years on end, so I was determined to find out what was actually wrong with me before doing anything stupid. I also wanted to find out the right way to start back up when the time was right. I didn’t want to rush into anything, but I could feel myself losing the exercise-every-day-after-work-and-go-to-the-pool-a-few-mornings-each-week habit that I had developed by the beginning of December.

On Friday, I went to my podiatrist, who said (again), “Boy, your feet are eff’ed the fuck up . .  all loosey-goosey and flat and shit.” And then he went on to say, “You don’t have plantar fasciitis, but you’ve gone and slightly fucked up the long tendon that connects your calf to your big toe via your heel. It’s amazing you’ve been able to get way without this kind of shit for so long. You need expensive orthotic shoe-inserts to keep this from happening again. Now, let me teach you some calf stretches and recovery techniques. You should start popping Aleve like a fiend, too. I’ll tape up your foot, and you can go running tonight if you want. But don’t go for any PRs or bullshit like that for a little while.” (I’m paraphrasing just a wee bit here.)

So I’m quite relieved. I’ve gone running twice since visiting my not-at-all-potty-mouthed podiatrist. Each run felt good, foot-wise. The left one isn’t 100% in the hours afterward, but it’s 10x better than the days after I injured myself. The runs also felt shorter and more difficult than I remember them being a month ago. Even so, these short, difficult runs were awesome.

Speaking of amazing things. I’ve been out on my road bike twice this new year already, and each time I wore shorts. New Years Day was the first time I’d been out since early October, and the lingering chill on the thawing roads couldn’t bring me down. Saturday morning’s sunny, 50°F, 25-mile ride had no chill at all. By way of contrast, at this point last year we had more than 30 inches of snow on the ground, and we were in for 60 more.

So I guess there’s that, too.

Oh, and there’s swimming! The Friday before Christmas I got up super-early despite not needing to go to the office. The pool was open, and I had the chance to get a full hour-and-a-half swim, instead of my typical 40-or-so minutes. The last time I had this opportunity, I swam two miles, and I wanted to give it another go, testing my blood glucose along the way. The results were very much like last time—better actually. My BG stayed almost constant; my 250-yard split times were fairly consistent throughout; and I swam a quarter mile farther in the same amount of time.

Now that I’ve written this, I’m reminded how fickle I can be. Yes, winter can be a cold, dark, lonely, depressing, snowy, stir-crazy-making time of the year. But it seems that all I need is a good report from the doctor, a run or two, an outdoor bike ride, a nice swim, and the constant loving support of Lisa for me to feel like a good spring is just around the corner.


p.s. I guess I should add that last night Lisa and I watched a documentary about U.S. athletes in the Beijing Olympics. It wasn’t the best thing ever, but it sure looked beautiful on our new high-def TV. I can barely wait to see this year’s games. Hurry summer don’t be late.

Posted in Cycling, Life Lessons, Reluctant Triathlete, Running, Swimming | 4 Comments