Microsoft Research offers tools and more at AGU 2013 Fall Meeting

Published

Every December I get together with 20,000 like-minded researchers in San Francisco to discuss how to preserve the habitat of Homo sapiens. I concede it’s a self-serving goal, but I’m okay with standing to benefit. Our conversation invariably burrows into subtopics of how the Earth works as a complex system because scientists agree: you can’t preserve a habitat until you understand it, and we need to make some progress on that front.

Helping scientists overcome big data challenges at the American Geophysical Union Fall Meeting

Enter Tony Hey (opens in new tab), brandishing the Fourth Paradigm (opens in new tab), a guidebook for working effectively in the burgeoning field of data-intensive science. I like to paraphrase the main premise of this new paradigm as follows: “Good for you, scientists, you’ve figured out how to get vast quantities of new data to help you better understand the Earth; but alas, you still need to build the analysis engines to actually make sense of the data!” Cue Microsoft Research and its contributions at the American Geophysical Union Fall Meeting in San Francisco from December 10 to 13. That’s the get-together I mentioned above.

Microsoft Research Blog

Introducing Aurora: The first large-scale foundation model of the atmosphere

Aurora, a new AI foundation model from Microsoft Research, can transform our ability to predict and mitigate extreme weather events and the effects of climate change by enabling faster and more accurate weather forecasts than ever before.

What will we—Microsoft Research—be doing on the trade show floor of San Francisco’s Moscone Center? Two things. First, we’ll be finding out what scientists are working on and what technical challenges they face in their work. Second, we’ll be offering them technology tools and resources to help them meet some of those challenges.

In that pool of technologies, one of our best contributions this year comes via the Windows Azure for Research (opens in new tab) project, a grant program that presents researchers with a year of free cloud-computing resources. Windows Azure, the Microsoft public cloud platform, offers scalability, immense computational power needed to analyze big data, redundancy, freedom from IT maintenance tasks, many virtual machine options including categories of Linux, and quick-install website templates to help them share out results.

At AGU, we will also present:

  • Layerscape (opens in new tab), a research toolkit built around the WorldWide Telescope virtual globe visualization engine. Layerscape enables anyone to harness the GPU power of a PC to render 10 million data points evolving in time in geospatial, solar system, or abstract reference frames.
  • FetchClimate (opens in new tab), a free service that quickly provides climate data and provenance across region, time, and data type parameters—from soil carbon to surface air temperature to precipitation and more.
  • DataUp (opens in new tab), a web application that helps researchers share, document, and archive their data (and receive Digital Object Identifiers in exchange for publishing).
  • Distribution Modeller (opens in new tab), a browser-based Bayesian inference engine for building ecological models.
  • Excel Power X and Office 365 (opens in new tab): The top 15 surprising/astounding things a researcher can do with Excel—from treating the web as a direct data source to using axis sliders for charts—and how Office365 is a powerful new way to get things done faster and with greater reliability.

We will also be explaining how we help teachers and scientists (opens in new tab) use and develop world-changing technologies through grants, fellowships, and internships. As always, we’ll be happy to share our ideas on data-system confederation, data publication, and data sharing. We will also provide short talks on selected geoscience topics such as estimating snowpack water storage in the Hindu Kush by using computers and remote sensing, and how to get your code running on a Linux Virtual Machine in the Microsoft cloud.

So that’s where we’ll be, from December 10 to 13, asking questions, explaining our tools and resources, and helping scientists make technology really work for them. And we love it. Nothing compares to the feeling of finding a scientist who needs something that we have and, even better, surprising him or her by explaining it’s available at little to no cost, aside from their time investment to learn the ropes.

If you plan to attend the AGU Fall Meeting, please stop by our booth. But even if you aren’t going to be in San Francisco, you can find an overview of the tools and technologies described above on the AGU events page (opens in new tab) on the Microsoft Research website. And while you’re checking out these resources, you might be tempted to ask (as curious scientists do), “What’s in this for Microsoft?” That’s the easiest question of all, because when it comes to trying to preserve the habitat of Homo sapiens, the benefits of success are self-evident.

Rob Fatland (opens in new tab), Senior Research Program Manager, Microsoft Research Connections

Learn more