Conflicted about HSAs....

I am having a very mixed reaction to the recent public announcements of AMD's Heterogeneous System Architecture (HSA) roadmaps. As anyone who stumbles across this blog should know, I did a lot of work in that area. I really do (or is it "did"?) think it represents a real future of computing silicon. It should reduce overall power and is the roadmap to create truly powerful single-chip computers that do everything. Building hotter FPUs or more cache or more generic cores is a race of diminishing returns.

We already have significant sections of modern x86 CPUs dedicated to special purpose functions. Vector instructions with SSE started us down the path, but now we have instruction sets and silicon that speed encryption and other low level functions. The advantages of this model is a single instruction set and, most importantly, a flat memory space.

It is the overhead of memory copy that actually eats up most of the advantage of dedicated silicon. The problem has to be large enough to be worth shipping around the data. Phil Rogers, Mark Hummel & others at AMD know this well. But so do the teams at NVidia, Intel and every other silicon company in the world.

Unfortunately, the silicon industry has also shown stamping out small power-efficient general purpose cores is easy. The difference in power consumption between on-load dedicated cores as NIC and purpose built silicon is shrinking. However, the design overhead of developing that silicon has not. There are only a handful of compute problems worth solving in silicon, but there should be 1000s of compute functions to which it is worth dedicating a core.

AMD has an approach to coherent memory across the different silicon environments. I know enough of the people involved to be confident the solution is elegant and functional. I have deep confidence that it can be game changing in the HPC space where uber-FLOPS still matter and adoption is a matter of compiling in the libraries.

Unfortunately, I don't think it is industry changing. Current software trends are not toward getting more from a single program, but dividing up the problem into smaller, general purpose compute elements. Software architects realized it was too hard to copy the data, so they moved the compute to the data. Yes, I'm talking about Map/Reduce.

AMD's HSA and Map/Reduce represent two directions for software's use of silicon. In my opinion, the need to speed up specific algorithms has been circumvented already by software architecture. AMD is shooting where the duck was, not where it is going. That is the problem with silicon engineering in the current age. It doesn't move fast enough.

That's why Dan Reed, Burton Smith and others are working for MSFT these days and Peter Ungaro is giving the keynote at a Big Data conference. They are trying to shoot ahead of the duck.

If AMD can improve memory efficiency (e.g. garbage collection) and/or messaging primitives (e.g. collectives) they may not change the industry, but you certainly have a competitive advantage in the modern age of distributed software architectures.

http://www.hpcwire.com/hpcwire/2012-02-09/amd_opens_up_heterogeneous_computing.html

Thinking about Start-ups?

Navin Thandani recently left Red Hat and decided to talk about the StartUp experience.

It's a classic tale of failure, woe & iteration, but with a happy ending. A couple of particularly nice points about the ease of getting someone to care about you.

http://nthadani.wordpress.com/2012/02/02/a-retrospective-analysis-on-the-road-to-red-hat/

Happy New Year

A belated New Year greeting. It has been a crazy year for accelerated computing and the deployment of special purpose hardware.

1. The HPC world clearly has embraced GPUs as a route to getting more Flops. It is unclear if any of it actually adds up to a lot more science, but the tools are in the hands of the experts.

2. Though we have many more cores, not a single new paradigm for using them has emerged. Libraries are being optimized for MPI on the system and interest in OpenMP has increased. These modest incremental updates are, in a word, useless.

3. Virtualization in the real world is there. With most jobs able to max out a single socket, virtualization of clusters is starting to make sense - not just for deployment scenarios, but also for job scheduling.

4. The problem right now isn't memory bandwidth. It's I/O. Big Data style solutions that move the compute to the data are going to be increasingly popular. However, isn't this what siesmic has been doing for years? What is really going to be new?

5. SSSDs are mainstream. But the reference architectures by use case are still being developed.

6. Rise of PaaS should create opportunities for more specialized deployments, similar to the combined software and hardware solution of Amazon's latest DB efforts. NoSQL on SSSDs... what next?

Thoughts for January. Tune in for more thoughts in February.

doug

What do Subcommittees Think?

Interesting(?) line up at Ben Quayle's next hearing...

Two EMC companies, a Microsoft and the GSA on Cloud Computing. Two Dr(s) and two Mr(s) in front a room of JDs (lawyers) talking about current wave of tech.

In all fairness, it should be noted that Capellas in addition to being CEO of VCE is also the chair of TechAmerica and the other Mr (David McClure) has been front and center in fedearl Cloud adoption & cost.

I can guess what they will say. It is the doctors who might add some interest to the event. I suspect that Nick Combs of EMC will make a product-like pitch. But Dan Reed of MSFT tends to be higher level and aspirational. He is one of the handful of very interesting technologists Craig Mundie brought on several years ago. They need to start delivering on something more interesting that "To the Cloud" soon.



Mr. Michael Capellas, Virtual Computing Environment Company

Dr. Dan Reed, Corporate Vice President, Technology Policy Group, Microsoft Corporation

Mr. Nick Combs, Federal Chief Technology Officer, EMC Corporation

Dr. David McClure, Associate Administrator, Office of Citizen Services and Innovative Technologies, General Services Administration


http://science.house.gov/hearing/technology-and-innovation-subcommittee-hearing-cloud-computing

Before Pandora...

Life in a cube is not one of quiet contemplation. It isn't one of privacy or focus.

I have an excellent set of in-ear isolation headphones. They don't help me get work done, but they do help me ignore others. I listen to Pandora mostly. I accept the advertising as a consequence of the pleasant surprises that sneak into my ear.

Feist's cover of the Beatles was perfect for avoiding sending that nasty email to a colleague on why the were (once again) failing to make my job easier.

Lately, I've been streaming other music on my iPhone (over Wifi) when I work out in the gym. It reinforced how little I care about the computer I use. I only care about the services that I can access.

Windows 8 is chasing the iPad/iPhone like Android is. Frankly, I know Android is better at some activities than my iPad or iPhone. There is nothing stopping Windows8 from being better than both at some tasks - so long as it remains open to also serve all the services I expect.

Could I become a Windows user again? I certainly don't want to go back any Windows that looks or acts like they did. All it has to be is a version of Windows that makes me forget that Windows ME, Clippy & Bob ever existed.

Maybe they should start with a new name?

Trying something new

My job(s) have evolved. I'm further from the tech these days & spending more time on purchasing behaviour, perception and reach. I'm also going to evolve this blog.

I tend to use Mondays to plan the week & check on progress toward strategic goals. That includes course correction on those goals. In that spirit, I also use Mondays to read material not exactly in my space. Here are my Monday factiods:

The economic impact of the "Office Lady" is global.

In Asia, the young, employed, stay-at-home female not only spends, but drives brand awareness - the alpha geek of fashinistas
http://bit.ly/ef30oy
Yet, Japan still has the largest gender gap in salaries.

In the US, young women out earn young men.
http://bit.ly/eXAsH5
And, despite some great women leaders, female CEOs are rare enough to make news

Men claim to be the primary grocery shopper
Ad Age reports that 51% of men claim to be the primary grocery shopper. And are more brand aware.

B2B marketing increasing is online
From email to virtual events to social media everyone is spending more online. This includes an increased investment in Social Media - mostly for "Brand Building". In other words, the impact of social media is something that can not be directly measured.
They also plan to hire.
http://bit.ly/h4P851

The Acceleration market is alive...

AMD's Fusion finally saw the light of day and it's getting awards. How this translates into better computing, more horsepower and better programming models is still to be seen.

http://www.zdnet.com/blog/computers/ces-2011-amds-fusion-apus-arrive-in-new-laptops/4656

With this hardware actually in the wild, I'm looking forward to hearing about students, professors and researchers actually programming something on it. If it only saves power, it is a sound silicon design. It's industry impact will be dependant upon simplified programming and more effective computing.
I haven't seen that demo-ed anywhere yet. Have you?


Meanwhile, the world sees a need to speed up math. Those GHz aren't getting any faster, are they?
http://www.hpcwire.com/features/Egyptian-Startup-Accelerates-High-Performance-Accounting-113319679.html