Hewlett Packard Labs and the power of parallel innovation
November 5, 2018 • Mark Potter, Chief Technology Officer and Director of Hewlett Packard Labs • Blog Post
IN THIS ARTICLE
- Labs is pushing HPE onwards in the race to exascale and a new age of supercomputing
- Memory-Driven Computing is helping HPE move the needle on innovation for a variety of computational challenges
- Hewlett Packard Labs is focused on scientific innovation, from mapping the known universe to mapping the human brain
Our focus on delivering technology breakthroughs leads to innovation – both expected and unexpected
For the last half century and more, Hewlett Packard Labs has been charged with fueling engineering innovation at HPE. Often working behind the scenes but always on the cutting edge of technology, Labs is part of the culture and DNA of our great organization.
In my role as Director, I follow in the footsteps of some incredible and inspiring leaders beginning with our founding director, Barney Oliver, and extending to my talented predecessor, Martin Fink. I have been honored to assume stewardship of a team and program that has continually strived to reach an ambitious goal: “Delivering technology breakthroughs that can change our world.” That is a very tall order!
Hewlett Packard Labs focuses on placing long-term bets that will propel us forward. We invest in smart technology and even smarter people — these are the critical assets required to develop the solutions that will shape tomorrow.
These extraordinary technology innovations may help scientists solve world hunger. Or map the human brain. Or take us all the way to Mars.
We live in a world where ideation and innovation have accelerated beyond what we ever could have imagined. Enterprises large and small are looking to leverage new technologies to capitalize on business opportunities and enhance humankind. HPE must join those redefining what is possible, to apply our innovations to solve our greatest problems.
The early Romans first proclaimed that “fortune favors the brave.” At Labs, we believe it favors those brave enough to take on, and achieve, the biggest missions of our time.
But just being brave and being the first is not enough. One must have a core belief in the journey undertaken; at Labs, that mission-based belief is our “true north,” a direction anchored by diverse ideas and inputs as we follow a serpentine path to the truth.
Some say Fleming discovered penicillin by mistake; or Arthur Fry’s Post-It Note was a lucky invention. Neither is true. Fleming spent thousands of hours, and filled hundreds of petri dishes with experiments, before unearthing the wonder of penicillin. And the Post-It Note was the byproduct of Fry’s countless attempts to uncover the impossible. Hewlett Packard Labs works in this same spirit. We aim to solve problems others deem to be out of reach, continually testing ourselves in the process.
Our challenges at Labs are big and aspirational, and we are privileged to tackle them. But we notice our process yields a dual-path pattern; some innovations are expected and predictable, while others occur seemingly by happenstance, and are the unexpected byproduct of our explorations.
Memory-Driven Computing falls into the latter category. As we built bigger and bigger computers, we were paralyzed by the law of diminishing returns. So, we learned from but ultimately broke free of 70 years’ worth of incrementalism, and applied the lessons learned and insights gleaned to ignite a new era of open innovation.
For example, our cooperation with German neurodegenerative disease research organization DZNE is helping them harness Memory-Driven Computing principles to garner the computational muscle needed to fight Alzheimer's and other seemingly unsolvable brain diseases.
As we worked on these and other critical missions, we reasoned that Memory-Driven Computing could win the race to exascale — a supercomputer that can complete a billion billion calculations per second (that’s 1,000,000,000,000,000,000 FLOPS). That means the computing power in one machine is the equivalent of the computing power currently spread across the world’s top 500 supercomputers. It’s not enough to use Memory-Driven Computing to simply compute more; we must do it with greater energy efficiency, too; otherwise, our “solution” may create even bigger problems.
As we work on these projects and challenges, one technological discovery often begets another. No world-changing or society-saving idea is so big, or too audacious, to be ignored. Because along the way we experience unplanned, “Post-It Note moments” that may have huge implications for the future of our planet.
Any Labs project may yield a great surprise. As our Chief Architect, Kirk Bresniker, says: Right now is the most exciting time for innovation. And the bigger the mission, the more we reveal being right in the middle of the fray.
We are creating technologies that may foster positive change for generations to come. The goal may be distant, and the journey long. But, oh, what a trip it can be!
Want to learn more? Check out these sessions at HPE Discover 2018 in Madrid, Nov. 27 – 29.
Do we have to wait for quantum computers?
Speakers: Marco Fiorentino, Rebecca Lewington
Thursday, November 29, 2018 from 12:30 PM-01:00 PM
There’s a new wave in computing needed to solve “impossible” problems in cryptography, scheduling, supply chain optimization and biochemistry. Quantum computing may eventually be able to crack these problems, but are there quicker, cheaper, more practical ways to reach many of the solutions that quantum computing is trying to provide? In this session, Distinguished Technologist Marco Fiorentino will demonstrate how Hewlett Packard Labs is thinking beyond the qubit, even beyond the transistor, to make the impossible possible.
How big can Memory-Driven Computing go?
Speakers: Mike Vildibill, Paolo Faraboschi
Thursday, November 29, 2018 from 11:30 AM-12:00 PM
At a billion-billion calculations per second, exascale computing will profoundly impact our ability to analyze, simulate and solve complex problems, arriving at solutions never thought possible. Scientists and businesses will need computers with great sophistication and power efficiency. Join HPE’s thought leaders to learn how a combination of technologies and systems expertise is accelerating us toward exascale to shape the next phase of economic and human development, and how that enables solving high-end computing problems in the enterprise.
Memory-Driven Computing: Composing the AI-driven future enterprise from edge to cloud
Speakers: Kirk Bresniker, Gary Thome
Tuesday, November 27, 2018 from 10:00 AM-10:30 AM
In a future where everything computes, intelligence will be embedded everywhere. Everyone and everything will be connected, driving exponential increases in data. Customers need to find new and better ways to distill data into meaningful insights and actions at speeds faster than the competition. Memory-Driven Computing is the solution to the limitations of conventional systems. Join HPE thought leaders to discover how our photonics technology and novel accelerators make Memory-Driven Computing the ultimate composable infrastructure.
What is the right hardware for (the next generation of) AI?
Speakers: Cat Graves, Ashkan Seyedi
Wednesday, November 28, 2018 from 02:30 PM-03:00 PM
Increased computational power of hardware is one of the key enablers of the recent revival of Artificial Intelligence (AI), but there isn’t enough computational power or energy to run an AI-everywhere future. This has propelled both industrial giants and small startup efforts to develop special purpose hardware for AI. Which approach will win? Join Hewlett Packard Labs researchers to learn how the requirements for AI algorithms are driving our unconventional approach to creating novel accelerators with unprecedented performance and efficiency.