DOCSLIB.ORG

Scientific Data Management at the Johns Hopkins Institute for Data

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Scientific Data Management at the Johns Hopkins Institute for Data Scientific Data Management at the Johns Hopkins Institute for Data Intensive Engineering and Science Yanif Ahmad, Randal Burns, Michael Kazhdan, Charles Meneveau, Alex Szalay, Andreas Terzis fyanif,randal,misha,[email protected], fmeneveau,[email protected] Johns Hopkins University, Baltimore MD 21218, USA. http://idies.jhu.edu 1. INTRODUCTION braries, Applied Math, School of Medicine, School Scientific computing has long been one of the of Public Health and the Human Language Tech- deep and challenging applications of computer sci- nology Center of Excellence. The authors of this ence and data management, from early endeavors paper make up the Computer Science members of in numerical simulation, to recent undertakings in IDIES and two of their long-standing collaborators. the life sciences, such as genome assembly. Complex The establishment of a data intensive computing computational problems abound and their solutions center has numerous benefits in addition to provid- transform our understanding of the physical world. ing an umbrella for data-driven research. It acts as The data management community's interest in sci- a beacon for the remainder of the Hopkins commu- entific applications has grown over the last decade nity and beyond, where our members are the first due to the commoditization of parallelism, dimin- port of call for anyone with big data problems from ishing system administration costs, and a search for academia, industry, and government agencies in the relevance beyond enterprise applications. vicinity and nationwide, such as the Johns Hop- Research in scientific computing raises non-tech- kins Applied Physics Laboratory, a not-for-profit nical challenges, such as overcoming the paucity of research lab housing 3,000 engineers and scientists. resources needed for experimentation, and estab- The Johns Hopkins University provides an ex- lishing a collaborative research agenda that fosters tremely conducive environment for these efforts. a mutual appreciation of problems, results in a con- The demographics of the university leads to an certed effort to develop software tools, and makes outward-looking faculty within departments. The all researchers successful in their respective fields. model of research being conducted through inter- In light of this, we report on a recently formed in- disciplinary centers is the norm rather than the stitute at the Johns Hopkins University to further exception (e.g. the Human Language Technology the interaction between computer science, and sci- Center of Excellence). This facilitates application- ence and engineering. We describe ongoing projects driven research, combining scientific domain knowl- at the institute and our collaboration experiences. edge with core computer systems research expertise. From an academic standpoint, the interdisciplinary model of IDIES percolates through to the students, 2. IDIES engaging students in an application and dataset- The Institute for Data Intensive Engineering and driven research agenda. Our agenda fosters a breadth Science (IDIES) was founded in April 2009 to bring of knowledge beyond data management and com- together data intensive computing in a variety of puter science, and encourages a hands-on approach science and engineering disciplines. The mission to system development. This results in tangible, us- of IDIES is to foster the interdisciplinary develop- able components that further the underlying scien- ment of tools and methods that derive knowledge tific problem, as well as promoting an experimental from the massive datasets that today's instruments, approach to data management challenges. experiments, and simulations generate at exponen- This research model leads to the development tially growing rates. of communities around individual datasets and the curation of these datasets both in terms of data 2.1 Research and Academic Model cleaning and functionality supported on top of the IDIES includes faculty members drawn from a dataset. Most often, this process produces public variety of disciplines ranging from Physics and As- resource for widespread academic usage. tronomy, Mechanical Engineering, the Sheridan Li- 2.2 Scientific Data Workflow at IDIES use SQL to formulate their research questions and Traditionally, under the scientific method, basic can run observations instantly, rather than waiting research involves the formation of hypotheses and for months to get their turn at a telescope. theories, designing experiments for their validation, The original database was built in close collabo- collecting data by realising experimentation, and ration between Alex Szalay's team at JHU and Jim analysing data to guide new insights for further re- Gray of Microsoft. The database is based on SQL search through iteration of this workflow. All stages Server, contains about 70 tables of data and de- of this workflow loop exhibit heavy computational scriptive metadata, and incorporates a substantial and data management needs, particularly experi- amount of astronomy code in the form of User De- mentation, simulation and analysis. The emergence fined Functions. Among other things, we have built of data-driven science has front-loaded much of this a high precision GIS system for astronomy, based on iterative workflow, where an intense period of ex- the Hierarchical Triangular Mesh, implemented as perimentation results in a substantially larger data a set of class libraries, wrapped into C# functions, volume for processing, followed by an intense pe- callable from T-SQL. riod of exploration and analysis of the data. Both Today about 30% of the world's professional as- modes must address large dataset challenges. tronomy community has a server side database in Many of these challenges appear individually in this system. The two-stage parallel loading envi- other domains, but are greatly exacerbated in scien- ronment and the whole framework is now in use tific computing by the diversity of complex, instance- by many groups, both at JHU and all over the specific semantics. The end result is an application world, and has been repurposed from astronomy domain with extremely high overheads and barriers to many other disciplines, such as environmental to entry through the combination of limited soft- science and radiation oncology. Most recently, the ware reuse, and the brittle infrastructures arising Pan-STARRS project, on the way to its first of from ad-hoc composition of tools that are difficult many petabytes of data, builds upon a scaled-out to parameterize and generalize. version of the SDSS design. We have assembled a critical mass of researchers at IDIES with expertise spanning many aspects 4. APPLICATIONS AND DATASETS of the scientific research workflow: Randal Burns IDIES research spans a wide variety of applica- (storage systems and transaction processing), An- tions and datasets. We present a sample here. dreas Terzis (sensor and wireless networks), Michael Kazhdan (computer graphics, mesh processing and 4.1 Turbulence geometric retrieval) and Yanif Ahmad (data man- Hydrodynamic turbulence is a formidably diffi- agement and declarative languages). We are ex- cult and pressing problem. However, direct numer- tremely fortunate to work with scientists and en- ical simulations of turbulence for the flow conditions gineers who have developed a deep appreciation of prevalent in most engineering, geophysics, and as- data management throughout their careers, includ- trophysics applications are impractical. Also, most ing Alex Szalay (physics, astronomy), and Charles critical scientific problems, such as predicting the Meneveau (mechanical engineering, turbulence). Earth's climate and developing energy sources, re- quire reliable simulations of turbulent flows. 3. THE SLOAN DIGITAL SKY SURVEY Typically, individual researchers perform large The focus on data-intensive computating at JHU simulations that are analyzed during the compu- started when the Department of Physics and As- tation with only a small subset of data stored for tronomy joined the Sloan Digital Sky Survey project subsequent analysis. The majority of the time evo- (SDSS), the astronomy equivalent of the \Human lution is discarded. As a result, the same simu- Genome Project." The SDSS created a high reso- lations must be repeated after new questions arise lution multi-wavelength map of the Northern Sky that were not initially obvious; most breakthrough with 2.5 trillion pixels of imaging. The results concepts cannot be anticipated in advance. Thus, a of the image segmentation were placed in a rela- new paradigm is emerging that creates large and tional database that has grown to 12 terabytes over easily accessible databases that contain the full the last ten years and has created a new way to space-time history of simulated flows. do astronomy. Based upon citation statistics, the In IDIES, we have embarked on building such database of the SDSS has been the most used as- databases. The JHU public turbulence database tronomy facility in the world. A large fraction of houses a 27 TB database that contains the entire the world's astronomy community has learned to time history of a 10243 mesh point pseudo-spectral sectors of the U.S. economy in terms of its energy consumption. According to a 2007 EPA report, U.S. data centers consumed 61 billion kWh in 2006| enough energy to power 5.8 million households. Un- der conservative estimates, IT energy consumption is projected to double by 2011. Only a fraction of the electricity consumed powers IT equipment. The rest is used by environmental control systems such
Load more

Recommended publications
  • Astrophysics with Terabytes
    Astrophysics with Terabytes Alex Szalay The Johns Hopkins University Jim Gray Microsoft Research Living in an Exponential World • Astronomers have a few hundred TB now – 1 pixel (byte) / sq arc second ~ 4TB – Multi-spectral, temporal, … _ 1PB • They mine it looking for 1000 new (kinds of) objects or more of interesting ones (quasars), 100 density variations in 400-D space 10 correlations in 400-D space 1 • Data doubles every year 0.1 2000 1995 1990 1985 1980 1975 1970 CCDs Glass The Challenges Exponential data growth: Distributed collections Soon Petabytes Data Collection Discovery Publishing and Analysis New analysis paradigm: New publishing paradigm: Data federations, Scientists are publishers Move analysis to data and Curators Why Is Astronomy Special? • Especially attractive for the wide public • Community is not very large • It has no commercial value – No privacy concerns, freely share results with others – Great for experimenting with algorithms • It is real and well documented – High-dimensional (with confidence intervals) – Spatial, temporal • Diverse and distributed – Many different instruments from many different places and many different times • The questions are interesting • There is a lot of it (soon petabytes) The Virtual Observatory • Premise: most data is (or could be online) • The Internet is the world’s best telescope: – It has data on every part of the sky – In every measured spectral band: optical, x-ray, radio.. – As deep as the best instruments (2 years ago). – It is up when you are up – The “seeing” is always great – It’s a smart telescope: links objects and data to literature on them • Software became the capital expense – Share, standardize, reuse.
    [Show full text]
  • Things You Need to Know About Big Data
    Critical Missions Are Built On NetApp NetApp’s Big Data solutions deliver high performance computing, full motion video (FMV) and intelligence, surveillance and reconnaissance (ISR) capabilities to support national safety, defense and intelligence missions and scientific research. UNDERWRITTEN BY To learn more about how NetApp and VetDS can solve your big data challenges, call us at 919.238.4715 or visit us online at VetDS.com. ©2011 NetApp. All rights reserved. Specifications are subject to change without notice. NetApp, the NetApp logo, and Go further, faster are trademarks or registered trademarks of NetApp, Inc. in the United States and/or other countries. All other brands or products are trademarks or registered trademarks of their respective holders and should be treated as such. Critical Missions Are Things You NeedBuilt On NetApp to Know About Big Data NetApp’s Big Data solutions deliver high Big data has been making big news in fields rangingperformance from astronomy computing, full motionto online video advertising. (FMV) and intelligence, surveillance and By Joseph Marks The term big data can be difficult to pin down because it reconnaissanceshows up in (ISR) so capabilities many toplaces. support Facebook national safety, defense and intelligence crunches through big data on your user profile and friend networkmissions and to scientific deliver research. micro-targeted ads. Google does the same thing with Gmail messages, Search requests and YouTube browsing. Companies including IBM are sifting through big data from satellites, the Global Positioning System and computer networks to cut down on traffic jams and reduce carbon emissions in cities. And researchers are parsing big data produced by the Hubble Space Telescope, the Large Hadron Collider and numerous other sources to learn more about the nature and origins of the universe.
    [Show full text]
  • Gerard Lemson Alex Szalay, Mike Rippin DIBBS/Sciserver Collaborative Data-Driven Science
    Collaborative data-driven science Gerard Lemson Alex Szalay, Mike Rippin DIBBS/SciServer Collaborative data-driven science } Started with the SDSS SkyServer } Built in a few months in 2001 } Goal: instant access to rich content } Idea: bring the analysis to the data } Interac@ve access at the core } Much of the scien@fic process is about data ◦ Data collec@on, data cleaning, data archiving, data organizaon, data publishing, mirroring, data distribu@on, data analy@cs, data curaon… 2 Collaborative data-driven science Form Based Queries 3 Collaborative data-driven science Image Access Collaborative data-driven science Custom SQL Collaborative data-driven science Batch Queries, MyDB Collaborative data-driven science Cosmological Simulations Collaborative data-driven science Turbulence Database Collaborative data-driven science Web Service Access through Python Collaborative data-driven science } Interac@ve science on petascale data } Sustain and enhance our astronomy effort } Create scalable open numerical laboratories } Scale system to many petabytes } Deep integraon with the “Long Tail” } Large footprint across many disciplines ◦ Also: Genomics, Oceanography, Materials Science } Use commonly shared building blocks } Major naonal and internaonal impact 10 Collaborative data-driven science } Offer more compung resources server side } Augment and combine SQL queries with easy- to-use scrip@ng tools } Heavy use of virtual machines } Interac@ve portal via iPython/Matlab/R } Batch jobs } Enhanced visualizaon tools 11 Collaborative data-driven science
    [Show full text]
  • Harnessing Grid Resources to Enable the Dynamic Analysis of Large Astronomy Datasets Year 1 Progress Report & Year 2 Proposal
    January 27, 2007 NASA GSRP Proposal Page 1 of 5 Ioan Raicu Harnessing Grid Resources to Enable the Dynamic Analysis of Large Astronomy Datasets Year 1 Progress Report & Year 2 Proposal 1 Year 1 Proposal In order to setup the context for this progress report, this section covers a brief motivation for our work and summarizes the Year 1 Proposal we originally submitted under grant number NNA06CB89H. Large datasets are being produced at a very fast pace in the astronomy domain. In principle, these datasets are most valuable if and only if they are made available to the entire community, which may have tens to thousands of members. The astronomy community will generally want to perform various analyses on these datasets to be able to extract new science and knowledge that will both justify the investment in the original acquisition of the datasets as well as provide a building block for other scientists and communities to build upon to further the general quest for knowledge. Grid Computing has emerged as an important new field focusing on large-scale resource sharing and high- performance orientation. The Globus Toolkit, the “de facto standard” in Grid Computing, offers us much of the needed middleware infrastructure that is required to realize large scale distributed systems. We proposed to develop a collection of Web Services-based systems that use grid computing to federate large computing and storage resources for dynamic analysis of large datasets. We proposed to build a Globus Toolkit 4 based prototype named the “AstroPortal” that would support the “stacking” analysis on the Sloan Digital Sky Survey (SDSS).
    [Show full text]
  • Warmth Elevating the Depths: Shallower Voids with Warm Dark Matter
    MNRAS 451, 3606–3614 (2015) doi:10.1093/mnras/stv1087 Warmth elevating the depths: shallower voids with warm dark matter Lin F. Yang,1‹ Mark C. Neyrinck,1 Miguel A. Aragon-Calvo,´ 2 Bridget Falck3 and Joseph Silk1,4,5 1Department of Physics & Astronomy, The Johns Hopkins University, 3400 N Charles Street, Baltimore, MD 21218, USA 2Department of Physics and Astronomy, University of California, Riverside, CA 92521, USA 3Institute of Cosmology and Gravitation, University of Portsmouth, Dennis Sciama Building, Burnaby Rd, Portsmouth PO1 3FX, UK 4Institut d’Astrophysique de Paris – 98 bis boulevard Arago, F-75014 Paris, France 5Beecroft Institute of Particle Astrophysics and Cosmology, Department of Physics, University of Oxford, Denys Wilkinson Building, 1 Keble Road, Oxford OX1 3RH, UK Downloaded from https://academic.oup.com/mnras/article/451/4/3606/1101530 by guest on 28 September 2021 Accepted 2015 May 12. Received 2015 May 1; in original form 2014 December 11 ABSTRACT Warm dark matter (WDM) has been proposed as an alternative to cold dark matter (CDM), to resolve issues such as the apparent lack of satellites around the Milky Way. Even if WDM is not the answer to observational issues, it is essential to constrain the nature of the dark matter. The effect of WDM on haloes has been extensively studied, but the small-scale initial smoothing in WDM also affects the present-day cosmic web and voids. It suppresses the cosmic ‘sub- web’ inside voids, and the formation of both void haloes and subvoids. In N-body simulations run with different assumed WDM masses, we identify voids with the ZOBOV algorithm, and cosmic-web components with the ORIGAMI algorithm.
    [Show full text]
  • Large Scale Structures of the Universe Proceedings of the 130Th Symposium of the International Astronomical Union, Dedicated to the Memory of Marc A
    springer.com Physics : Astronomy, Observations and Techniques Audouze, J., Pelletan, M.-C., Szalay, A. (Eds.) Large Scale Structures of the Universe Proceedings of the 130th Symposium of the International Astronomical Union, Dedicated to the Memory of Marc A. Aaronson (1950–1987), Held in Balatonfured, Hungary, June 15–20, 1987 Ten years ago in August 1977 Malcom Longair and Jan Einasto organized IAU Symposium nO 79 on exactly the same exciting and most important topic i.e. the Large Scale Structure of the Universe. Many of us have the recollection of an outstanding meeting which fulfilled two goals (i) establish most of the foundation of a fast growing field (ii) set up a confrontation between the excellent observational and theoretical work performed in eastern and western countries. A decade after such a meeting Alex Szalay and I have felt the need to reassemble the Springer cosmologists working actively on problems dealing with the Uni• verse as a whole. Indeed a 1988, XXXIV, 622 p. lot of progress has been achieved in the building of large surveys in the discovery of voids, 1st sponges and filaments in the galaxy clus• ter distribution, in refined numerical simulations, in edition experimental and theoretical particle physics (outcome of new particles (cold particles) and unification (GUT, supersymmetry) schemes), in research of quantum gravity and inflation scenarios etc ... A new confrontation between all the specialists working all throughout the Printed book world on such questions appeared to us to be most timely. This is why the location of Hardcover Balatonfiired in Hungary to be accessible to anyone as Tallin in 1977 has been chosen.
    [Show full text]
  • 2016 – Aug – 09 CURRICULUM VITAE Mike Specian Contact
    2016 – Aug – 09 CURRICULUM VITAE Mike Specian Contact Information: Home Address: 7023 Heathfield Rd.; Baltimore, MD 21212 Mobile: (443) 653-1690 Email: [email protected], [email protected] Webpage: www.mikespecian.com Twitter: @mspecian Education: 2009-2015 – Ph.D., Johns Hopkins University, astrophysics Dissertation: Improved Galaxy Counting Techniques and Noise Reduction Algorithms as Applied to the Sloan Digital Sky Survey, advisor: Prof. Alex Szalay 2004-2009 – M.A., Johns Hopkins University, astrophysics 1999-2003 – B.A. cum laude and with distinction, Boston University, joint concentration in physics and astronomy Professional History: 2016-present Visiting Scientist – Johns Hopkins University – same Work as When I Was an assistant research scientist, but With a different title 2016 Guest Instructor, Beth Tfiloh Dahan Community School – taught classes to prepare high school students for the AP Physics exam 2015 Assistant Research Scientist, Johns Hopkins University – research focused on making corrections to the Sloan Digital Sky Survey footprint, mapping galaxies Without redshifts, applying noise filters to galaxy overdensities 2012-2013 Scientific Consultant, US Global Change Research Program’s National Climate Assessment; Washington, DC – provide recommendations of physical climate change indicators to the technical team 2012-2013 Member, Media and Message Working Group for MD House of Delegates Campaign; Baltimore County, MD – conduct background research on House of Delegates members, write Op-Eds, contribute to campaign strategy
    [Show full text]
  • Migrating a Multiterabyte Archive from Object to Relational Databases
    H IGH-D IMENSIONAL D ATA MIGRATING A MULTITERABYTE ARCHIVE FROM OBJECT TO RELATIONAL DATABASES A commercial, object-oriented database engine with custom tools for data-mining the multiterabyte Sloan Digital Sky Survey archive did not meet its performance objectives. We describe the problems, technical issues, and process of migrating this large data set project to relational database technology. he advent of digital archives—en- avoid being overwhelmed by the flood of data. abled by quantum leaps in the tech- The Sloan Digital Sky Survey (SDSS) is a multi- nology to publish, distribute, and institution project to map about half of the north- mine data over the Internet—has ern sky in five ultraviolet-to-infrared wavelength createdT a science and engineering data avalanche. bands (www.sdss.org). Projected for completion in The Human Genome Project (www.ornl.gov/ 2005, the survey should image over 200 million ob- TechResources/Human_Genome/ home.html) jects and collect spectra (redshifts) for the bright- and the CERN Large Hadron Collider (http:// est one million galaxies among the objects. A de- lhc-new-homepage.web.cern.ch/lhc-new- tailed description of the SDSS project appeared in homepage) are two examples of very large sci- a previous issue of CiSE.1 To learn more about the entific data sets coming online in the biologi- SDSS science archive (SA) discussed in detail here, cal and particle physics communities, respec- go to www.sdss.jhu.edu. tively. Astronomy is no exception, and soon SDSS will revolutionize astronomy in many ways, might be deluged with more new data from but most significantly, it will dwarf current astro- current and proposed sky surveys than any nomical databases.
    [Show full text]
  • Extreme Data-Intensive Scientific Computing on Gpus
    Extreme Data-Intensive Scientific Computing on GPUs Alex Szalay JHU An Exponential World 1000 • Scientific data doubles every year 100 – caused by successive generations 10 of inexpensive sensors + 1 exponentially faster computing 0.1 2000 1995 1990 1985 1980 – Moore’s Law!!!! 1975 1970 CCDs Glass • Changes the nature of scientific computing • Cuts across disciplines (eScience) • It becomes increasingly harder to extract knowledge • 20% of the world‟s servers go into centers by the “Big 5” – Google, Microsoft, Yahoo, Amazon, eBay • So it is not only the scientific data! Scientific Data Analysis Today • Scientific data is doubling every year, reaching PBs • Data is everywhere, never will be at a single location • Architectures increasingly CPU-heavy, IO-poor • Data-intensive scalable architectures needed • Databases are a good starting point • Scientists need special features (arrays, GPUs) • Most data analysis done on midsize BeoWulf clusters • Universities hitting the “power wall” • Soon we cannot even store the incoming data stream • Not scalable, not maintainable… Non-Incremental Changes • Science is moving from hypothesis-driven to data- driven discoveries Astronomy has always been data-driven…. now becoming more generally accepted • Need new randomized, incremental algorithms – Best result in 1 min, 1 hour, 1 day, 1 week • New computational tools and strategies … not just statistics, not just computer science, not just astronomy… • Need new data intensive scalable architectures Continuing Growth How long does the data growth continue?
    [Show full text]
  • The World-Wide Telescope, an Archetype for Online Science
    The World-Wide Telescope, an Archetype for Online Science Jim Gray, Microsoft Research Alex Szalay, Johns Hopkins University June 2002 Technical Report MSR-TR-2002-75 Microsoft Research Microsoft Corporation One Microsoft Way Redmond, WA 98052 1 2 The World-Wide Telescope, an Archetype for Online Science Jim Gray Alex Szalay Microsoft Research The Johns Hopkins University [email protected] [email protected] Abstract: Most scientific data will never be directly statistics among sets of data points in a metric space. Pair- examined by scientists; rather it will be put into online algorithms on N points scale as N2. If the data increase a databases where it will be analyzed and summarized by thousand fold, the work and time can grow by a factor of a computer programs. Scientists increasingly see their million. Many clustering algorithms scale even worse. instruments through online scientific archives and analysis These algorithms are infeasible for terabyte-scale datasets. tools, rather than examining the raw data. Today this analysis is primarily driven by scientists asking queries, but The new online science needs new data mining algorithms scientific archives are becoming active databases that self- that use near-linear processing, storage, and bandwidth, and organize and recognize interesting and anomalous facts as that can be executed in parallel. Unlike current algorithms data arrives. In some fields, data from many different that give exact answers, these algorithms will likely be archives can be cross-correlated to produce new insights. heuristic and give approximate answers [Connolly, Astronomy presents an excellent example of these trends; Szapudi]. and, federating Astronomy archives presents interesting challenges for computer scientists.
    [Show full text]
  • Mark C. Neyrinck University of the Basque Country Curriculum Vitæ Mobile Ph: +34 663 490 872 (Spain) 48080 Bilbao, Spain +1 808 232 7263 (US)
    Ikerbasque Fellow [email protected] Dept of Theoretical Physics Mark C. Neyrinck http://origami.science University of the Basque Country Curriculum Vitæ Mobile ph: +34 663 490 872 (Spain) 48080 Bilbao, Spain +1 808 232 7263 (US) Born in Colorado, USA Principal Achievements - I developed engaging “origami” and “cosmic spiderweb” descriptions of cosmic structure formation, which helps to understand information loss in the cosmos, and how angular momentum in a galaxy is related to its environment, leading to involvement in a NOVA program - I pioneered techniques to detect and analyze cosmic voids for cosmological constraints. With my premier void-finder ZOBOV, we made the first detection of the cold imprints of voids on the cosmic microwave background, a sign of dark energy. - I discovered a simple way to enhance the power of conventional large-scale-structure studies: Gaussianizing transforms. This will lead to a deeper understanding, and tighter constraints, on cosmic structure formation and the cosmological and galaxy-formation physics driving it. Popular Media Attention for Work “Cosmic filaments may be the biggest spinning objects in space,”, Science News, 22 June 2021 “Art of the Cosmos: Understanding galaxies through origami,”Artful Science interview, Jun 2021 Interview, SciArt Magazine, August 2020 “The Cosmic Web that Connects Galaxies Together May Be Spinning," New Scientist, 16 June 2020 Paper-folding, dark matter and the structure of the universe, Science Gallery Dublin podcast, Apr 2020 Interview and segment about origami and cosmology in NOVA episode “The Origami Revolution,” https://rmpbs.pbslearningmedia.org/resource/buac17-912-sci-ess-nvtorcosmicfold/wgbh-nova-the-origami- revolution-cosmic-folding/ “The Cosmic Spiderweb on Dark-Matter-Haloes’ Eve” invited blog post, The Huffington Post “The Origami Cosmic Web” article, The Paper (OrigamiUSA magazine, circ.
    [Show full text]
  • The Importance of Data Locality in Distributed Computing Applications Alex Szalay, Julian Bunn, Jim Gray, Ian Foster, Ioan Raicu
    The Importance of Data Locality in Distributed Computing Applications Alex Szalay, Julian Bunn, Jim Gray, Ian Foster, Ioan Raicu Current grid computing environments are primarily built to support large-scale batch computations, where turnaround may be measured in hours or days – their primary goal is not interactive data analysis. While these batch systems are necessary and highly useful for the repetitive ‘pipeline processing’ of many large scientific collaborations, they are less useful for subsequent scientific analyses of higher level data products, usually performed by individual scientists or small groups. Such exploratory, interactive analyses require turnaround measured in minutes or seconds so that the scientist can focus, pose questions and get answers within one session. The databases, analysis tasks and visualization tasks involve hundreds of computers and terabytes of data. Of course this interactive access will not be achieved by magic – it requires new organizations of storage, networking and computing, new algorithms, and new tools. As CPU cycles become cheaper and data sets double in size every year, the main challenge for a rapid turnaround is the location of the data relative to the available computational resources – moving the data repeatedly to distant CPUs is becoming the bottleneck. There are large differences in IO speeds from local disk storage to wide area networks. A single $10K server today can easily provide a GB/sec IO bandwidth, that requires a 10Gbit/sec network connection to transmit. We propose a system in which each ‘node’ (perhaps a small cluster of tightly coupled computers) has its own high speed local storage that functions as a smart data cache.
    [Show full text]