Hyperscaling The engineer of the 21st century


While January is the month of predictions for many, I have made it a habit to look back and see how previous forward-looking assessments have worked. It’s fascinating to see how many predictions in the past were wrong and how little has changed in some areas.

Twenty years ago, in January 2001, the front page of IEEE Spectrum put the theme of ubiquitous connectivity in an ever-active world. Some of the predictions in “Welcome to the always-on world” from Philip E. Agre during his time at UCLA were terrifyingly accurate. Agre writes that:

“We will increasingly be freed from geographical constraints and equipped with powerful search tools, we will be able to select exactly the people we want to interact with and we will be able to connect with them at any time. We will of course not fall into disembodied brains, and geographic proximity will always play an important role in our lives. Rather, the point is that we can have more continuous relationships with anyone with whom we associate near or far. “

Agre predicts a future where email will be checked anywhere, anytime, which has been a reality for quite some time. As mentioned earlier, security will be of paramount importance in the relevant consumer, cellular and network industries.

With regard to the transport sector, “Elisabeth A. Bretz” is “The car: just a web browser with tires“Gives a great view. Scott McNealy is quoted as saying, “Why not just look at everything as an Internet hub, so a car is just one [web] Browser with tires. ”Today, 20 years later, we achieve this goal with almost complete connectivity. At the time, the article focused heavily on “component manufacturers who work overtime developing Bluetooth-enabled cars”. What deserves more and more attention nowadays is the interoperability of old and new “devices”, ie cars, on the same road network in so-called “Long Lifecycle Markets”. A full quarter of a century ago, in January 1996, IEEE Spectrum’s transportation division focused on how satellites will replace microwave landing systems, how faster, smarter ships shimmered on the horizon, how planes could get help against micro-explosions, and how – back then – Texas led the way with intelligent highways .

The same 1996 issue has an article on the aerospace and defense industries that talks a lot about development costs. The considerations on competitiveness from the point of view of “being ahead of the competition – and staying ahead” by Don Fuqua, the then President of the Aerospace Industries Association, still seem modern today. Some things stay the same. Fuqua also says that “after about a decade of dedicated pursuit of new efficiencies, many companies have exhausted traditional approaches to cutting costs and increasing productivity as well as anything they can. To take competitiveness to the next level, approaches such as “flexible manufacturing” are required. ”He later defines this as“ extending computer-aided design and manufacturing (CAD / CAM) to include the manufacture of a part or component from scratch computerized tools to include. In a broader sense, there are competitive players – costs, quality, reliability and less development time – as important in design considerations as the performance of a product. “

Last year Will Roper, the Air Force’s Deputy Secretary for Procurement, Technology and Logistics, presented a draft for “Disruptive agility for a disruptive world”Also focused on cost, avoiding re-rotations and re-application. Likewise, The Laws of Augustine came out in 1986 and stated, “In 2054 the entire defense budget will buy just one tactical aircraft. This aircraft must be shared by the Air Force and Navy each three and a half days per week, except during the leap year when it is made available to the Marines for the extra day. So it is fair to say that the topic of costs was, is today and will be the main topic in this industry in the foreseeable future.

What does all of this mean for us engineers? We need to hyperscale for the era of hyperscale computing!

In the January 2001 issue of IEEE Spectrum, article by Joseph Bordogna “The engineer of the 21st century” identifies five skills that will shape the future of engineering: terascale, nanoscale, complexity, cognition and holism.

Terascale would take us “three orders of magnitude beyond the current universal and generally accessible computing capacities”. In 2001 the first Apple Store opened, the release of MAX OS X and Windows XP, the first Microsoft XBOX and the first release of iTunes (Source: Museum of Computer History). Twenty years later we have a cloud-based, scalable availability of computing.

Nanoscale “Will put us three orders of magnitude below the size of most man-made devices today.” Okay, no question about it. Ten years ago I wrote about us that we were “discussing 16nm / 15nm technology nodes”. Today we are discussing 3nm and beyond, as well as “More than Moore” techniques that enable us to continue to grow in complexity. Speaking of which, Bordogna’s article referred to complexity to the point at which “the components of a system never fully engage and yet never completely dissolve into turbulence”.

holism is “the concept that a unit is more than just the sum of its parts”. Bordogna concludes that “the hallmark of the modern engineer is the ability to see connections between seemingly different components and integrate them in ways that exceed the sum of their respective capacities”. One could argue that although disciplines are increasingly interconnected – software and hardware, electrical and mechanical, etc. – engineers who can understand all areas and become experts are still rare. Instead, communication and interfaces between tools and of course the use of abstraction have become central and will become even more so in the future.

While the absolute numbers we have achieved over the past two decades may be controversial, all of these deliberations in direction have been remarkably precise. Scale itself is key, and it will be interesting to see its sequel in a hyperscaled world. And as for the changing requirements, as some of the challenges have been so imminent for decades, Quote by William Gibson from 2003 seems to apply today more than ever: “The future is already here – it is just not evenly distributed!”

Frank Schirrmeister

Frank Schirrmeister

(All posts)

Frank Schirrmeister is Senior Group Director for Solutions and Ecosystems at Cadence. He leads a team tasked with translating customer challenges in the hyperscale, communications, consumer, automotive, aerospace / defense, industrial and healthcare verticals into specific requirements and solutions. His team focuses on cross-product technical solutions such as 5G, artificial intelligence, machine learning, safety, security and digital twins as well as on important partner collaborations.

Source link


Leave A Reply