To Compete You Must Compute
Thème : L'infrastructure numérique
Côte de l'idée : +463 | Total de votes : 499 | Commentaires : 40
High Performance Computing (HPC) or Supercomputing is critical digital infrastructure for productivity, innovation and research and for building and sustaining a digital economy, indeed, a digital society. In the 21st century, countries that want to compete will have to compute. This will require:
- Sustained and state–of–the–art supercomputing infrastructure in order that Canadian researchers remain globally competitive and the private sector investment in R&D at Canada's academic institutions results in world–class innovation and competitive advantage in the marketplace.
- The development of the supercomputing (HPC) professionals — Highly Qualified Personnel — within academia that are required by the private sector to ensure innovation and productivity improvements.
- Continued support for academic research utilizing supercomputing and recognition of its contribution to economic prosperity and the health and well–being of Canadians.
- Positioning Canadian researchers in both academia and in the private sector at the forefront of technological change in supercomputing.
- Encouraging academic–private sector research partnerships for using HPC for research and for HPC research.
- The establishment of core funding for supercomputing infrastructure to ensure that Canada becomes a leader in this field and that we are able to attract and retain top talent capable of effectively utilizing this resource.
Canadian researchers depend upon the availability of competitive high performance computing facilities in order to participate in addressing economic, social, and medical challenges in collaboration with researchers in Canada and around the world. If our researchers do not have the tools necessary to support their participation, Canada and Canadians will lose. We will not be the innovators we need to be in a digital economy nor will we be able to effectively address some of the key challenges facing Canadians. At great expense, both economically and socially, we will be adopters rather than leaders; reactive rather than proactive. And we will continue to lag in productivity relative to those countries with whom we compete.
Encouraging Canada's small and medium sized enterprises to adopt and adapt supercomputing for competitive advantage and sustainability will give them leverage in the marketplace. Compute Canada has signed a Memorandum of Understanding with the State of Montana's Rocky Mountain Supercomputing Centers to undertake an initiative to enable participation and competitive advantage in the digital economy for Small–to–Medium–sized Enterprises by providing access to High Performance Computing facilities and expertise. The issue for Compute Canada is how to fund Canada's participation to ensure that our SMEs are digital economy ready. As stated by the United States Council on Competitiveness in the 2008 Report "The New Secret Weapon": "Supercomputing is part of the corporate arsenal to beat rivals by staying one step ahead of the innovation curve."
Computers have re–invented everything we do. Supercomputers are re–inventing how research is conducted and accelerating the timeframe for that research. Some businesses are already using HPC to advantage. The movie industry uses HPC for animation and rendering of special effects; retailers regularly use data mining; and credit card companies use HPC for fraud detection. Simulations to design chemicals that help the immune system fight bacteria now take 3 months on one of Compute Canada's new supercomputers rather than the more than ten years it would have taken without it. The Canadian academic environment and Compute Canada are prepared to work with the private sector through investment in R&D to realize such order of magnitude impacts on the Canadian economy.
High Performance Computing, or Supercomputing, is a very specific component of Canada's digital infrastructure. It is the critical vehicle for creating knowledge from the vast amount of data being collected or generated. In an era of sophisticated modeling of complex systems and the necessity of analyzing massive datasets, it is vitally important to understand the importance of HPC to a modern knowledge–based economy. If modern optical networks are the 21st century equivalents of roads and railways, then HPC corresponds to the major cities that generate and use the products that flow on these transportation arteries… and that produce the wealth of nations. Compute Canada is leading the creation of a powerful national HPC platform for research. This national platform integrates High Performance Computing resources at seven partner consortia across the country to create a dynamic computational resource.
noskovsy — 2010–07–08 12:24:13 HAE a écrit
This is an excellent suggestion. HPC does change the way we think about research. All of a sudden we can do myriads of things at the same time unraveling fundamental principles hidden in large arrays of data. Access to HPC is critical to so many areas of our life spanning from pure science such as astrophysics and particle physics to weather modeling, genome/drug research to day–to–day banking.
progan — 2010–07–08 12:51:02 HAE a écrit
To take full advantage of investments in HPC infrastructure, it will be necessary to create state of the art fiber optic data infrastructure between HPC centers and within the intranets of the organizations that use these machines. Increasingly, HPC requires migration of large datasets between machines or across networks.
ahmedb — 2010–07–08 13:37:51 HAE a écrit
Hardware is fundamental. But what's hardware without the software. I believe increasing efficiency needs some kind of a strategy for software management/sharing. Obviously licensing (if applicable) needs to be worked out.
renew — 2010–07–08 13:45:07 HAE a écrit
HPC is at the center of my genomics research. Without it, we would NOT be able to process, analyze and make sense of the volume of sequencing data we generate daily. We currently rely on a in–house, state of the art computing infrastructure. In the near future that may not be sufficient, as the data generation becomes more affordable — at a rate that supersedes that of the computer hardware costs. Creating a national computational resource is a step forward; I'm delighted that such an initiative may be implemented in our country.
mewhort — 2010–07–08 15:35:05 HAE a écrit
I am an academic psychologist who studies how human memory works. The work has both theoretical and practical implications. Applications include knowledge retrieval and translation. Without HPC, the field is dead in Canada, and my students will continue to contribute to the US economy. With HPC, my students can stay in Canada.
Ron Van Holst — 2010–07–08 15:56:04 HAE a écrit
HPC technology has been commoditized, so there has never been a time to invest in HPC as a competitive differentiator. GPU technology is being added to HPC, promising orders of magnitude greater performance, a real discontinuity in computing power, as long as associated software can be adapted to take advantage of it.
Yes, SME's need to adopt HPC technology, in this case the needed scale of hardware is relatively inexpensive today, the barrier here would be the cost of software and the expertise required to use the numerical power to solve problems or create better solutions for these enterprises. Things like CFD solvers to design better machines.
But Canada should invest on a national scale, for a petaflop machine for grand challenge engineering and science problems. Things like climate modelling and bioinformatics research. Looking at the Top 500 supercomputers, and the relative size of Canada's economy, we should already be building such a machine. Canada should also participate with international forums on the development of exaflop systems.
Jasmin E Roy — 2010–07–08 15:59:43 HAE a écrit
HPC is a primary concern in my research. I carry out Finite Difference Time Domain (FDTD) simulations of electromagnetic wave propagation/scattering/radiation with an in–house developed software with unique capabilities. The range of applications is widespread, from design/analysis of antennas, photonics components, electromagnetic interference, metamaterial, etc. HPC has opened new possibilities in my research. I would not think of doing without it now.
TerryDalton — 2010–07–08 16:03:04 HAE a écrit
HPC as long been realized through many international initiatives as a means of increasing competitiveness. Canada needs to embrace and promote what has been developed through Compute Canada. We have to provide appropriate access to this Canadian infrastructure asset (cyberinfrastructure) to the research, innovation and educational community. That means access even to the SME innovation engine of Canada which makes up the highest collective economic driver we have. Access can be provided through the advanced digital infrastructure Canada has developed by organizations such as CANARIE Inc. in partnership with the Provincial Optical Regional Advanced Networks such as Atlantic Canada Organization of Research Networks in Nova Scotia (ACORN-NS), RISQ, BCNET, ORANO, ACORN-NL, MRnet, SRnet, Cybera, ECN, and many other organizations. Critical to making this sustainable is the commitment to increase the investment in developing the skill set needed in Canadians in HPC and Digital Infrastructure… invest in the people to stimulate creative and curiosity thinking for our future… invest in our kids.
joezli — 2010–07–08 16:15:12 HAE a écrit
In modern world, scientific computing is an indispensible tool to make productive and quality research.Yeah, to compete we have to compute with powerful high performance computers and highly efficiently software as well as well–educated experts. Canada needs computing power to compete with the nations. Joe
sergeychelsky — 2010–07–08 16:17:42 HAE a écrit
HPC is a key to success in any scientific research these days. Unfortunately, Canada is way behind all other developed countries in this competition. I vote that it should be changed.
TraceyLauriault — 2010–07–08 16:27:11 HAE a écrit
This is awesome! Also, could some of you vote to ensure we have content flowing through those pipes! Open Access to Canada's Public Sector Information and Data
Open Access to Canada's Public Sector Information and Data
zhaoga — 2010–07–08 16:39:10 HAE a écrit
Without HPC most of the problems we solve are of the nature of conceptually idealized scenarios only. HPC makes massive simulation possible, which allow us the capacity to deal with real field problems.
mdharsee — 2010–07–08 17:28:58 HAE a écrit
In genomics/proteomics/bioinformatics research HPC pervades at multiple levels — from storing and crunching large volumes of raw data generated from next generation sequencers and mass spectrometers, to enabling data integration and data mining of these datasets, to providing security and privacy with respect to confidential clinical data. This infrastructure is critical to retaining our highly qualified talent right here in Canada, and in attracting international business and investment in our biotech and research sectors. The need for HPC — the hardware, the software, and the people — will only continue to grow, particularly in light of the "grand challenge" of Personalized Medicine, which is as much a computational problem than anything else, given the complexity of the molecular mechanisms of biological systems and diseases. We have an opportunity to be at the forefront of this research, and investment in HPC will be critical to getting there.
mmonagan — 2010–07–08 17:51:12 HAE a écrit
In my oppinion, the need for a national HPC facility is overstated and an expensive solution. Today, one may purchase a 4 core desktop for under $2000 and a 16 core server for under $20,000. Very soon we will have a 16 core desktop for under $4000. These machines will be adequate for most HPC users. What we need is software expertise, not more computing cycles. The cost of building HPC centers is tens of $millions. The cost of getting new equipment every year is in the $millions. It would be better if we DECENTRALIZE our hardware hardware resources and share resources at a departmental/university level. If we centralize resources, then what will happen is that 1% of the users will suck up 90% of the cycles and the casual users will have to wait in batch queues for days to get some cycles.
jmpoutissou — 2010–07–08 18:36:19 HAE a écrit
One aspect of computing that is changing the economics of product development is the use of extensive simulation instead of prototyping (The Boeing 787 needed only 4 test wings instead of 18 for the 747). Simulation is becoming the key factor in getting a competitive advantage. Extensive simulation requires large scale computing: that means large scale computing infrastructure BUT ALSO and more importantly top level programming specialists. One cannot do without BOTH. Because of these essential requirements,developing countries cannot compete with their cheap manpower on that innovation front, hence a clear advantage for countries developing a computing literate workforce (here I am not referring to window users but advance programming specialists). This is one area where Canada can best exploit its university trained students.
dguptill — 2010–07–08 20:17:02 HAE a écrit
Having seen ace–net in action, I have concerns about usage and responsiveness.
Usage: It is a truism that any resource that is free will eventually be abused. Unless there is some payment–for–use mechanism (and I realize this is non–trivial to implement; there are many ways to do it), the HPC resources may be abused.
Responsiveness: Without the payment–for–use, HPC system administrators are limited in what they can, and cannot, do by their budget. That budget may not respond well to changing user requirements.
olinart — 2010–07–08 21:07:54 HAE a écrit
I strongly support the desirability of a national HPC facility based on my experience with the regional facility in Western Canada. The project I worked with could have opted either to purchase a dedicated computer cluster for its analysis or to use Westgrid which was then just being commissioned. In fact we could have used it effectively a year before it came available. Compared to a dedicated facility, the regional facility was able to deal much better with the peaks and troughs in our requirements. Equally important was the professional staff administering and maintaining the equipment and software. Centralization of this work yields very significant efficiencies.
pollard — 2010–07–08 22:43:21 HAE a écrit
Investment in HPC infrastructure whatever maybe the technology is critically predicated on long term, sustained funding. Without this, our youth and future scientific leaders in Canada will face a very uncertain research landscape on which to build their careers. For example, it takes about 15 years to transform a young, inspired undergraduate into a professor, with tenure, at a university. So their research, which will depend on the third leg of the stool of 21st century science, must be readily available in Canada (the other two legs are physical experiments and theory). HPC physical infrastructure and human support personnel, along with the computational scientists and engineers in our tertiary educational institutions serve to train students who will become the innovators of tomorrow. The C3.ca Long Range Plan for HPC (Engines of Discovery: the 21st Century Revolution) should be required reading by all who engage in the Digital Economy.
GeoffreyRockwell — 2010–07–09 06:53:23 HAE a écrit
Well supported HPC infrastructure and associated human support is important also if Canada is to develop its digital content industry (from animation to games to electronic texts.) HPC facilities will be the backbone to large–scale archives, large–scale content mining, and multimedia content sharing.
kmacneill — 2010–07–09 09:25:57 HAE a écrit
Increased HPC investment, to support SME sector, can provide an immediate advantage to Canada. If we look at one sector, renewable energy, and more specifically wind, an SME with access to an HPC to assist in the selection of a wind farm site, can make a site selection that generates 25%–50% more power than a site selected without access to HPC's. The result is a competitive SME, HQP job growth, and a greener economy.
pfcuttle — 2010–07–09 11:49:35 HAE a écrit
HPC is necessary for scientific research. Even if you can buy cheap and powerful desktops or small–scale servers, you still can not attain the speed–up of a real, large–scale cluster. It is an entirely different hardware category, honed and tuned for hi–speed numerical processing. We can not allow ourselves to lag behind in the scientific community by not contributing to efficient and nation–wide HPC.
AndreQuenneville — 2010–07–09 15:41:09 HAE a écrit
"If our researchers do not have the tools necessary to support their participation, Canada and Canadians will lose." — Well said. Straight up and to the point. Too few Canadians are even aware of how important this is.
dsanden — 2010–07–09 15:58:42 HAE a écrit
CUT or EXPAND
CUT — if it's just for 300 elite Compute Canada scientists to have fun, publish papers and advance their academic careers
EXPAND — if general canadians can have access — for example through local science centres, so taxpayer/voter kids can get an early and solid taste of the kind of fun Compute Canada researchers experience daily, and — Outliers style* — go on to become leading scientists themselves.
* Donnée aberrante (livre)
supercomputingisimportant — 2010–07–09 16:00:41 HAE a écrit
High performance scientific computing allows researchers to pursue questions that cannot be answered with experiments. Simulations can yield important insights and discoveries in a wide range of fields from biological science to astrophysics. Canada must support such fundamental research.
cgeroux — 2010–07–11 13:09:54 HAE a écrit
All of the comments I have read thus far, as well as the title of the main article carry the idea that to "compete [with other nations] we must compute" and while I believe being competitive with other nations is very important it misses the point.
At present we are faced by global issues that our planet has never had to deal with in the past, namely a species that has such a large impact on our planet. We have technologies that pollute soils and oceans, emit large amounts of gasses into the atmosphere, and change habitats on scales never before seen in the history of our planet.
While this might seem a bit doomsday–ish, or overstated, we must put things into perspective. Consider the impact that a single human makes on the earth from the requirements for shelter, clothing and food and that these requirements are only likely to grow as more developing nations aspire to the standards of the developed nations. Then consider the rate of population growth coupled with the already large preexisting population. Finally consider the fragility of the environment and how very interdependent species are. Removing one part of a habitat has very complicated affects on the rest of the habitat with larger more complex forms of life usually feeling the brunt of these losses.
This is not merely a matter of being competitive, environmentally–ethical, or even maintaining a comfortable environment for us to live in. This is a matter of long term survival for our species. In order to ensure the survival of our species we must increase our knowledge and understanding of the natural world. In turn allowing us to increase, or at least maintain, the ability of the earth to support us. HPC while not appearing to be directly related to these issues plays a very important role in our ability to make informed decisions about our impact of the environment. HPC allows researchers to tackle research problems not previously possible, and to process and understand the large amounts of data crucially necessary for making informed decisions about our complex natural world.
DominicLam — 2010–07–11 20:45:02 HAE a écrit
There is a general misconception that HPC is another hi–tech tool for researchers and/or commercial concerns. For those who benefit from using HPC can testify that in many occasions, it actually interacts with the users intellectually similar to a research colloeague and directly contributes to creativity and innovation. This is the most significant reason for supporting a well funded and well run HPC national facility.
johnw — 2010–07–12 08:08:57 HAE a écrit
I believe the Federal Government's investment in HPC over the last decade, primarily through the Canada Foundation for Innovation, has transformed the research environment within Canada. This is the case not only for researchers whose principle interests and expertise are in areas that are closely associated with, and dependent upon, HPC, but also for many "non–traditional" users of HPC. Access to a well supported, stable computational environment has encouraged many researchers to incorporate large scale numerical modelling and simulation into their work, allowing them to compete and collaborate more effectively with researchers in other countries.
As the first generation of researchers trained using these resources move out into the private sector Canada will begin to realise the significant economic benefits of this investment. It is important to keep in mind that what constitutes today's HPC will very quickly become the mainstream computing environment for the many of the SME's critical to Canada's economic future.
That said there is still a long way to go to build the resources and develop the expertise required to sustain and advance progress in this key area. To be internationally competitive it is essential that the Federal Government, in partnership with the provincial governments, the research community and industry, commit to the continued growth of HPC, building on the impressive and transformative accomplishments of the last decade.
hades — 2010–07–12 10:22:44 HAE a écrit
Computing on massive scales is becoming a game–changing technology in increasingly many areas of research, government and industry.
This goes beyond the traditionally compute–intensive research areas of science and engineering, and now also includes large–scale computing applications that have emerged in new fields such as internet technology (web search, social network management, …), business and finance (simulation for product development, derivatives pricing, supply chain management, optimal pricing strategies, …), government policy (electronic health records, emergency response simulation, …), and the biomedical area (large–scale medical image processing, …).
A sustained investment in large–scale computing infrastructure will allow Canadian Universities and Research Centres to:
- Educate a new generation of large–scale computing professionals, who will lead the way in making Canada successful in the compute–intensive ventures of the future,
- Develop new algorithms and methods for more efficient large–scale computing research and large–scale data–processing, which will enhance research breakthroughs and industrial competitiveness,
- Compete internationally in terms of research results, product innovation, and bright ideas and people.
ljdursi — 2010–07–12 11:46:15 HAE a écrit
It's important to realize that pushing forward Canadian supercomputing — which by most measures is absolutely bottom of the G8 in what we're willing to spend on it — is important for far more than helping our leading scientists.
Don't get me wrong; if it was only that, it would still be worth it; funding supercomputing is like building a laboratory that any Canadian researcher can use — for biology or physics, for music research or for analyzing historical economic data. Pushing forward our cutting edge research is crucial for innovation and development.
But it's more than that. For instance, it's vital to business. You can imagine that it's useful for designing airplane wings or car engines, but its use is much broader. In the US, tire manufacturers and even shampoo manufacturers use computer–powered modeling to enormously cut down the development cost of new products and get them to market faster. Many businesses — most famously WalMart, but now many others — use supercomputing to analyze the torrents of data their businesses produce to do their work better — to design better shipping routes, or plan store placements (or even items within a store) better. This new field of business analytics can make use of the hardware, software, and expertise generated through endeavors like this one.
The people who built Yahoo, Google, and Facebook and their data centres were by and large trained in the academic supercomputing world, and the next generation of web–entrepreneurs will be, too. To train these people, and broaden and deepen our digital skills and entrepreneurial sectors, we need the hardware, and also (as pointed out by several commenters) the resources to develop new software and expertise to use these systems — all of which are fundamental to this idea.
I wholeheartedly vote yes!
agdelma — 2010–07–12 12:05:59 HAE a écrit
Without access to HPC resources, my research would simply not be possible. Unfettered availability of state–of–the art computational technology is an essential component in retaining excellent scientists in Canada.
jinzishuai — 2010–07–12 13:26:22 HAE a écrit
This is no doubt one of the key areas where Canada is lacking behind the world in terms of advancing science and technology. If we want to change Canada from a traditional resource provider to a leading technology provider, HPC or generally ICT needs to play a more critical role in the country's daily life.
couchman — 2010–07–12 13:48:08 HAE a écrit
As others have pointed out, wherever data needs to be analysed or processes modelled or simulated — whether these data or processes arise from nature, society or the economy — computation is necessary. In an era of massive and growing datasets and of increasingly fine grained and complex processes, high–performance (or super–) computing is critical. Being able to compete internationally in business and research and being able to maintain a healthy society and vibrant, advanced culture will require Canada to confront the supercomputing need head on. The technology is relatively straightforward but it needs to be sustained and we need to ensure that the training is in place to enable it to be effectively integrated into the Canadian landscape.
rmawhinn — 2010–07–12 13:53:39 HAE a écrit
HPC is a valuable research tool, and having access to such a tool provides Canadian researchers with the necessary platform for developing world–leading and cutting edge technology.
gweiler — 2010–07–12 14:02:29 HAE a écrit
HPC research does not just solve academic problems. The tools and techniques being developed can help make HPC more accessible to business and industry to help keep Canadian enterprise competitive in the global economy. Colleges in particular can help apply HPC to practical problems.
wskoczen — 2010–07–12 14:14:41 HAE a écrit
HPC is valuable for innovation and research. While decentralized systems are important and adequate for many researchers, large scale centralized systems are indispensable in providing access to those who otherwise wouldn't be able to afford it. Access to stable, well supported computing infrastructure promotes the growth of traditionally not compute intensive research and provides good platform to involve and educate a new generation of professionals.
sgsinclair — 2010–07–12 14:42:58 HAE a écrit
Canada has made significant and wise investments in HPC. Continuing to invest in HPC will ensure the value of past investments and continue to enable leading research as well as economic opportunities. HPC is enabling a range of new, large–scale research in the digital humanities as we continue to try to better understand human culture and communication in a digital age.
SFOShea — 2010–07–12 16:55:11 HAE a écrit
High performance infrastructure (computing, networks, data storage, and middleware) is becoming even more important as the amount and complexity of data increases. We need infrastructure for research and for development of highly qualified people to meet needs not yet evident to most Canadians.
There is a significant issue of slow uptake in Canadian SMEs, but that must be dealt with at a management level and by reducing the barriers to entry for SMEs to use these tools.
sykes — 2010–07–12 23:18:55 HAE a écrit
HPC is an integral part of many disciplines now that span numerous areas in the sciences, engineering, and most recently, the humanities. Academic researchers see the value in HPC and so do numerous private companies. Every major computer manufacturer and software company support HPC. In fact, they are all working on ways to further integrate HPC into their corporate strategic direction. HPC needs the support of local and national platform funding agencies in a continuous funding model. HPC is not only about the infrastructure–highly qualified personnel is vital to the overall success of any HPC service. With sufficient support Canadian researchers will continue to be world class leaders.
datalibre.ca · Consultation (real) Results — 2010–07–14 10:17:46 HAE a écrit
[…] votes — To Compete You Must Compute, submitted by Susan Baldwin
2010–07–08 08:32:20 HAE, Susan Baldwin (Ex Officio) Executive Director […]
digEcon scandals « Oh! Canada — 2010–07–17 03:35:52 HAE a écrit
[…] To Compete You Must Compute with +463 votes A discourse on the importance of Supercomputers, it also appears to be an advertisement for Compute Canada […]