National Aeronautics and Space Administration

IT Talk July - September 2018 Volume 8 • Issue 3

The Big Data Wave

www.nasa.gov IT Talk In this Issue Jul - Sep 2018 Volume 8 • Issue 3

Office of the CIO NASA Headquarters 300 E Street, SW Message from Washington, D.C. 20546 3 the NASA CIO Chief Information Officer Renee Wynn Editor & Publication Manager Eldora Valentine Graphic & Web Designer Michael Porterfield NASA Ames CIO Coauthors Copy Editor 4 Award-Winning Paper Meredith Isaacs

Using Advanced Data IT Talk is an official publication of the Analytics To Maximize Office of the Chief Information Officer 6 Value of Voice & Text Data of the National Aeronautics and Space Administration, Headquarters, Washington, D.C. It is published by the OCIO office for all NASA employees and external audiences.

For distribution questions or to Automated Data suggest a story idea, email: Tagging Project [email protected] 8 To read IT Talk online visit: www.nasa.gov/offices/ocio/ittalk

For more info on the OCIO: v www.nasa.gov/ocio v inside.nasa.gov/ocio (Internal NASA network only) Office 365 v www.nasa.gov/open/ 11 www.facebook.com/NASAcio Message from the NASA CIO

Every hour, day, and year, NASA gathers millions of data-points (a.k.a Big Data) from our active missions. The collection rate is growing by leaps and bounds, handling, storing, and managing these data is a massive challenge. Our proprietary data is among our most valuable assets, and its strategic importance in our research and science is enormous. NASA is committed to making our data as accessible as possible, both for the benefit of our work and for the betterment of humankind. In this issue, we’ll explore why it matters how the Agency uses big data. NASA also relies heavily on big data to help the Agency with artificial intelligence (AI). By applying AI and machine learning, satellites control these systems seamlessly, making real-time decisions without awaiting instruction. We’ll take a look at how AI is helping NASA engineers to develop software and algorithms to meet the specific requirements of missions. And soon, NASA will be migrating to Microsoft Office 365, offering a breadth of capability to our users. The suite allows employees access to e-mail, calendar, contacts, and tasks through a Web browser on any authorized NASA computer with a PIV badge or Agency Smart Badge. I’m excited about this initiative because this technology is industry-leading for our workforce. This cloud-based e-mail, calendaring, and collaboration service supports NASA’s plans to enhance our Agency’s security posture. Finally, this quarter, we are pleased to highlight some prestigious industry honors earned at the Centers. I’m so proud of all the hard work happening right now. We are making a difference in leading the transformation of IT at NASA! So I hope you enjoy reading this issue!

CIO Center Visits. Clockwise from top: CIO Face-to-Face at Stennis Space Center, April 10-12; with Langley Research Center’s Acting CIO Catherine Prohaska and NASA SAISO Mike Witt; and with Federal CIO Suzette Kent and NASA Shared Services Center CIO Danny Harvill.

NASA OCIO IT Talk Jul - Sep 2018 3 NASA Ames CIO Coauthors JPL IT Directorate Award-Winning Paper Recognized By Penny Hubbard, Communication Specialist, Ames Research Center with CIO 100, Ames Research Center (ARC) CIO Jerry Davis, plored, cybersecurity considerations have been with Krishna Sampigethaya of United Tech- largely missing. This paper bridges this gap Computerworld nologies Research Center and NASA’s Parimal and addresses UTM cybersecurity needs and Kopardekar, took first place in the “Best of issues. It contributes a comprehensive frame- Best Places To Track” category, and second place overall, at work to understand, identify, classify, and as- April’s Integrated Communications Navigation sess security threats to UTM. Promising threat Work in IT Honors and Surveillance (ICNS) Conference. The con- mitigations, major challenges, and research ference focuses on providing an understanding directions are discussed to secure UTM today By Whitney Haggins, IT Communica- of domestic and international programs, imple- and into the future. tion Strategist, Jet Propulsion Labora- mentation strategies, standards development, tory, California Institute of Technology research, and ICNS technologies. Congratulations to Krishna, Parimal, and Jerry! They have coauthored an award-winning paper NASA’s Jet Propulsion Laboratory’s Their paper, “Cyber Security of Unmanned and established significant research, policies, (JPL’s) IT Directorate was recently rec- Aircraft System Traffic Management (UTM),” and processes to secure UTM and sUAS for the ognized with two industry honors. provides a comprehensive glimpse into Small benefit of society. Unmanned Aircraft Systems (sUAS) and the On May 1, JPL was named a 2018 CIO compelling impacts to society as they become Read the award-winning paper: 100 Honoree by CIO magazine. The CIO mainstream. Simply put, the millions of small https://intranet.share.nasa.gov/arc/ 100, now in its 31st year, honors 100 Unmanned Aircraft Systems, along with other it/Documents/2018-04-ICNS/Cyber- organizations that have distinguished aircraft of various shapes and capabilities, will security_of_UTM_Paper.pdf themselves through their innovative use soon be flying at low altitudes in urban environ- of technology to create business value. ments—and it is critical to ensure that they fly View their Integrated Communications Naviga- This is JPL’s seventh consecutive ap- safely, predictably, and efficiently. tion and Surveillance (ICNS) presentation: pearance on the CIO 100. The honorees https://intranet.share.nasa.gov/arc/ will be celebrated at a gala during the While research into the challenges, inherent to it/Documents/2018-04-ICNS/Cyber- CIO 100 Symposium in Palos Verdes, CA, UAS Traffic Management (UTM), are being ex- sec_UTM_ICNS_2018_Presentation.pdf August 13–15, and they will be profiled in the August issue of CIO magazine.

On June 18, JPL was named one of the 100 Best Places to Work in IT by IDG’s Computerworld magazine and ranked number 19 among large companies. This is JPL’s sixth consecutive appearance on the prestigious list, a ranking of the top 100 work environments for information technology professionals by Computer- world magazine, now in its 25th year. The list is compiled based on a compre- hensive questionnaire regarding compa- ny offerings in categories such as ben- efits, career development, training, and retention. In addition, Computerworld conducts extensive surveys of IT work- ers, and their responses factor heavily in determining the rankings.

Jim Rinaldi, JPL CIO, said: “We are ex- cited and honored to again be recipients of these industry honors. We strive to provide IT that matters and appreciate that we can present some of the excit- NASA Headquarters’ Information Technology and Communications Division (ITCD) held “A Day ing things we are doing and receive rec- with ITCD” on Wednesday, June 20. They shared IT and communications products and services, ognition. It’s truly gratifying to know that with on-the-spot consultations, demos, and workshops. Photo Credit: NASA/Garrett Shea we have a work environment where our people enjoy what they are doing.”

4 www.nasa.gov An Agency-Wide Approach to Software Asset Management By Kellie White, Communications Specialist, Marshall Space Flight Center

The Office of the Chief Information Of- Shared Services Center (NSSC) to the Audit prevention ficer (OCIO) is taking a proactive ap- Applications Program (AP) Office. The • Reduced frequency of audits, or proach to the challenge of software as- AP was assigned the task to implement of negative outcomes from audits set management. an effective Agency-wide Software As- • Compliance with Government set Management process and appoint- Accountability Office (GAO) NASA needs a comprehensive strategy ed Jane Maples to lead the effort. While and Office of Management and and approach to address the full soft- the concept of software management Budget (OMB) guidance on ware life cycle for both new and exist- is not new to any enterprise, it is a diffi- software license management ing software licenses for cloud and non- cult undertaking for an agency the size cloud platforms, including non-licensed of NASA, with numerous geographical Most everyone would agree that the or free software. The foundation for locations and process owners across benefits to be gained are significant, this approach was established with the multiple directorates. but realizing those benefits will take NASA Interim Directive (NID) 7150.113, substantial work. How do we get to the Software License Management, issued Fundamental to beginning the process benefits? The seven steps in the soft- in 2017. The Enterprise License Man- and achieving success, is understand- ware license life cycle are depicted in agement Team (ELMT), established in ing the problem. Currently, any software the picture below. 2008, provides another key compo- deployed within the Agency, regardless nent of the overall strategy. Now, we of its function, injects risk into NASA’s are prepared to build on the foundation infrastructure from a cybersecurity by establishing, regularly tracking, and and software compliance perspec- maintaining a comprehensive inventory tive, which has the potential to cost of NASA software licenses. NASA millions per year in waste, inef- ficiency, and noncompliance fees. To Who is responsible for managing soft- decrease this risk, NASA must deploy ware at NASA? Everyone who uses an Agency-wide approach to software software at NASA has a responsibility management that provides visibility and to ensure that the software they use or efficiency throughout the software life purchase is cycle, beginning with planning and con- tinuing through retirement and disposi- 1. approved for use by NASA, tion. Knowing the Agency’s inventory Utilizing a stakeholder-inclusive ap- 2. secure, and of software provides multiple benefits, proach with representation from multiple 3. being used in accordance with some of which are listed below. Centers and disciplines, the resulting the terms and conditions of the two-year implementation of this strat- license agreement. Cost avoidance/savings egy will provide the Agency with a cen- • Increased opportunity to reduce tralized, standardized, streamlined life- What software does NASA need to software cost and avoid paying cycle process for managing software track? significantly higher prices that delivers service to the customer in • Maximization of software a timely manner and that is automated 1. Commercial-off-the-shelf (COTS) investments without purchasing to the greatest extent possible. We will software and software license new products and support have greater insight into the software types (e.g., perpetual licenses, • Decreased costs for new projects entering and existing in NASA’s environ- software maintenance, term and maintaining existing systems ment; a comprehensive software inven- licenses, subscriptions, shareware, • Cost avoidance on audits tory; the ability to track and maintain software as a service [SaaS], and cyber threats software licenses, maintenance, and open source, freeware, and cost; an improved NASA security pos- demonstration/trial software that Cyber threat detection ture; and the elimination of unnecessary is intended to be deployed within • Increased awareness of cyber costs, duplication, and waste. NASA that may not have an initial threats to NASA’s data (i.e., acquisition cost) and mission and personnel) If you have questions related to the 2. Modified commercial-off-the-shelf • Quick removal of legacy IT Agency Software Lifecycle Manage- software (MOTS). hardware/software, which ment approach developed by the AP, decreases threat possibility please contact the Agency Software The Agency Software Manager (ASM) • Identification and mitigation of Manager (ASM), Jane Maples, via e- role recently transitioned from the NASA outdated/unsupported software mail at [email protected].

NASA OCIO IT Talk Jul - Sep 2018 5 Using Advanced Data Analytics To Maximize the Value of NASA’s Voice and Text Data By Justin Gosses, Data Scientist, Johnson Space Center; Anthony Buonomo, Data Scientist; An- drew Adrian, PhD, Data Scientist; Brian Thomas, PhD, Data Scientist, NASA Headquarters

NASA has vast amounts of data stored as text (ML) are helping to meet this need, and we using the data itself. The resulting model can or audio. Oftentimes, gaining insight into these are now poised to unlock the value of NASA’s approximate complex behavior that would, us- data requires humans to read or listen to many vast amounts data. NLP, a subfield of machine ing traditional approaches, be very difficult and files and summarize the results. This process learning, is concerned with how to program expensive to model. ML can be applied to data is expensive and, therefore, precludes many computers to analyze large amounts of natural- of many types, including raw text, text extracted potentially useful types of analyses simply be- language data, including both human speech via speech-to-text, or a subset of the whole text cause these analyses are too costly, too boring, and written text. If you have ever used Google extracted via NLP. There are many different ML or too time-consuming. Considering that we will Translate, created a reminder with Siri, or used algorithms that one might choose to apply to never be able to hire enough folks to handle the grammar checking in Microsoft Word, you have solve a given problem, and it is not always clear avalanche of data that already exists, let alone interfaced with an NLP program. which of the possible choices will be the best the new data that are coming in year after year, solution. we need to find a better solution. The amount of data that NLP will make available is daunting, but ML provides us with the means When applying both NLP and ML, a trained data Fortunately, recent advances in natural-lan- to understand all of this new information. ML scientist can help. Our team of data scientists guage processing (NLP) and machine learning allows us to teach or program the computer has been working with partners here at the

6 www.nasa.gov Agency to apply these technologies to help Research Access rate. This outstanding accuracy rate is a result solve NASA mission problems. We have iden- One of our text-based natural-language pro- of ML—the process of teaching a computer to tified avenues for speeding up analyses and cessing projects is building a service to auto- find statistical patterns in data without being improving the accuracy of business processes, matically recommend keywords in NASA Scien- explicitly programmed to do so. By utilizing ML and we have pointed out ways to reduce costs tific and Technical Information (STI) documents. in our development, this solution will enable while enabling our workforce to focus on high- us to save thousands of dollars in storage and er-order tasks. In this article, we highlight a few The STI program would derive great value out transfer costs as we switch to preserving an of our projects. of a good keyword recommendation system. increasing amount of sent and received e-mail. Why? Because they are tasked to organize and Speech-To-Text make accessible a wealth of NASA research. Moving Forward Speech recognition has a long history. Al- The standard mechanism for doing so is by When domain experts and data scientists though people have sometimes been disap- tagging keywords to papers. There are upwards work together to tackle a problem, we can pointed when the technology fails to live up to of 25,000 official STI NASA concepts, many of create solutions that free up time and mental human-level expectations, we have found that which only apply to small/niche areas of the energy, enabling our workforce to focus on building good speech-to-text projects is often overall collected work. Moreover, this official higher-impact tasks. Looking forward, many an exercise in matching each speech-to-text set only scratches the surface of the NASA of the projects in this space have applications technology’s strengths and weaknesses to the concept space. A few individuals cannot pos- wider than the teams and use cases they were requirements of the use case. We have found sibly understand and summon to mind all of originally built for. If NASA could enable more that key criteria when picking a speech-to-text the keywords that apply to the broad wealth of sharing of the labeled training datasets, meth- engine include whether computation occurs knowledge being produced at NASA. ods, code, and IT infrastructure for executing in the cloud, on an intranet, or on a local ma- these types of projects, it would pay back many chine; whether the tool is geared for keyword By providing an ML model to automatically times over. Toward this end, our Agency Data tagging, command and control, or transcription recommend tags, we can (at least) provide an Analytics Team has been working to understand tasks; accuracy during the use case; and how initial filter so that the folks who are assigning the needs for advanced data analytics at the important is it that we are able to update the tags have some help. For clarity, this system Agency and to present the case for the adoption speech model with specific background noise (or one like it) is not set to replace people; it of needed techniques (such as ML and NLP) or NASA-specific words. is only set to augment their workflow. It is an and the development of the requisite strategy, assistant—a “bionic arm,” if you will, for the policy, and technical infrastructure that will help Proof-of-concept applications we have devel- stewards of STI. drive us toward a vision of a future NASA where oped include adding voice control to a Web site workers can all leverage these benefits in a with a real-time map of the International Space This work is poised to free up time for these cost-effective manner. Station’s (ISS’s) current position. Speech-to-text stewards and allow them to direct their mental on a Web page represents a new user interface energy toward tasks that only they can do—the that might be useful when hands are busy or careful curation of a great source of informa- using a mouse and keyboard is inconvenient. tion in order to educate, spread knowledge, and Currently at NASA WebAssembly, a way to convert C code to Ja- reduce duplication of research. vaScript, was used to create a system where all and want to get speech-to-text computation happened inside E-mail Records Classification the local browser. This approach paves the way Since 2008, NASA has collected and archived involved? for building small speech-to-text applications over 11 million e-mails from certain “Capstone” without requiring lengthy security reviews, officials. These e-mails contain a mix of Federal Check out these working groups: elevated privileges to install software, or the records and non-records that must be sorted security implications of sending data between before these e-mails can become part of the Science Mission Directorate and OCIO’s computers or outside the network. The data do public record. For example, the subject “Boost- Big Data Working Group: contact John not move outside the control of the user, and no er Thrust Test Results” would likely be a record, Sprague ([email protected]) software is installed. whereas “Late to work” would likely not be a record. The manual reading and sorting of these Artificial Intelligence Community of Another proof-of-concept in this area involved 11 million messages is estimated to take some Interest: contact James McClellan writing code to scan through spacewalk train- 23 years of human effort—and by the end of ([email protected]) ing videos from the Neutral Buoyancy Labora- that time, we would have many more e-mails tory and create data visualizations of tagged to sort! Therefore, we require an automated, Aeronautics Research Mission keywords. Having a visual representation of “smart” solution for this problem. Directorate’s Data Analytics Community the clustering of key terms linked back to their of Interest: contact Yuri Gawdiak place in the video sped up searching in what Our team has developed a smart solution that ([email protected]) would normally be a human-intensive task of reads and sorts NASA business e-mails “au- listening to the videos and taking notes. tomagically,” with up to a 96 percent accuracy

NASA OCIO IT Talk Jul - Sep 2018 7 Capitalize On Data Management, Automated Data Tagging Project Access, And Innovation By David Kelldorf, JSC Chief Technology Officer, and Truong Le, Data Architect, NASA’s IT Strategic Plan – Goal 2 Johnson Space Center By Jonathan Walsh, IT Strategic Planner, and Meredith Isaacs, Communications Specialist, NASA Last year, Johnson Space Center (JSC) developed Headquarters the first Center-wide IT planning activity and cre- ated the JSC Master IT Plan (available at https:// Data and information are cornerstones of NASA’s past, present, and future missions and discover- inside.nasa.gov/ocio/ocio-news/what-will-it-jsc- ies. The Agency’s 2018 Strategic Plan depends on a robust capability to conduct research and look-2037), which aligns with the 20-year JSC interpret data in order to accelerate decision making and discoveries. NASA’s IT Strategic Plan Center Plan. From this planning vision, three plan- for Fiscal Years 2018-2021 provides the focus needed for the Agency to manage IT as a strategic ning goals emerged to guide the development resource to securely unleash the power of data. The use of data to amplify the benefits of NASA’s process: a collaborative and evolving IT environ- missions and stewardship requires the ability to transform NASA’s data into insights and increase ment, secure and accessible data, and a flexible secure data accessibility through innovation. workforce.

To achieve this strategic goal, NASA will establish Agency-wide data standards supported by se- To support the first two goals, the Information cure, integrated data search and analytics capabilities to accelerate insights and improve the qual- Resources Directorate (IRD) at JSC initiated the ity of decision making across NASA’s complex data-rich environment. We will also investigate new Automated Data Tagging Project in June 2018. In technologies, take risks on disruptive ideas, and invest in operationalizing prototypes aligned with this project, a prototype is underway to investigate mission needs to increase secure data access and usability. mechanisms to automatically identify and catego- rize unstructured data. This initiative is designed Through data management, the use of analytics, modernized infrastructures, and collaboration, to make unstructured data discoverable and ac- NASA will remain a leader in Federal data sharing (open.nasa.gov; data.nasa.gov), increase data cessible and to identify high-value assets (HVA). availability and mobility, and innovate with projects such as financial data visualizations, extrave- hicular activity (EVA) suit data management, and scientific research. The current state of NASA data presents challeng- es in securing data, engineering security controls, NASA’s IT Strategic Plan outlines the Agency’s IT goals, objectives, strategies, and metrics for Fiscal and managing data overall. The strategic goals of Years 2018 – 2021. The plan focuses NASA’s IT community on strategic IT priorities, such as data the Automated Data Tagging Project that address management, accessibility, and innovation, that support achievement of NASA’s 2018 Strategic these challenges include simplifying discovery Plan and its missions. The Agency uses IT to share NASA’s results and deliver strategic IT capabili- and use for authorized users while providing the ties that drive discovery while increasing quality, productivity, mission safety, and cost optimization. appropriate degree of security and control rel- evant to data categories, empowering data-driven Check out NASA’s IT Strategic Plan here: https://www.nasa.gov/ocio/itsp strategic decisions and intentional IT operating Questions? - [email protected] model choices, right-sizing storage costs while enhancing monitoring and management, and de- veloping data categories that align with Agency and network data loss prevention tools.

The Automated Data Tagging Project will develop data models and connectors for file shares and SharePoint storage. Testing will focus on IRD- owned data shares and SharePoint repositories as a microcosm of the JSC data challenges. The goal is to discover and unify the view of unstructured data and develop strategies to store, share, and tag the information.

Partnering with the Agency team that has been developing tools for the Extravehicular Activity (EVA) Office, IRD will test developed tools for ap- plicability on corporate unstructured data.

The Automated Data Tagging Project will investigate mechanisms for HVA and provide recommendations This visualization of NASA’s IT Strategic Plan highlights for JSC and Agency architectures. The outcome of how the plan emphasizes data and information. The more this project has the potential to scale to other NASA a word was used, the larger and bolder it is seen here. Field Centers, making NASA data more discoverable and accessible while driving space exploration.

8 www.nasa.gov Big Data Big Think @GRC By Terry Jackson, Associate CIO for Technology, Data & Innovation Division, NASA Headquarters

NASA held a Big Data Big Think at Glenn Re- Chief Information Officer (CIO) Sean Gallagher search Center (GRC) from May 22 through 24. and other GRC representatives. Experts from A very diverse group of subject matter experts around the Agency discussed science and en- (SMEs)—including members of the Aeronautics gineering projects, high-performance comput- Research Mission Directorate (ARMD), Human ing and analytics, IBM Watson projects, Internet Exploration and Operations Mission Directorate of Management Things (IoMT), scientific work- (HEOMD), Space Technology Mission Director- flows, secure code sharing, and NASA World- ate (STMD), Office of the Chief Engineer (OCE), Wind. Special Lightning Talks were conducted, Attendees at NASA’s Big Data Big Think meeting Centers, and many others from across NASA— giving programs and Centers ten minutes to in Cleveland, OH. Photo by Herb Schilling, GRC were in attendance at this joint meeting spon- share best practices; OCE also helped lead a sored by the Office of the Chief Information data roadmap discussion. The keynote address Officer (OCIO) and the Science Mission Direc- was given by Ted Friedman, a Gartner data ex- torate (SMD). This was the sixth gathering since pert. During the conference, participants were the first one held at Langley Research Center treated to a visit to the Cleveland Museum of Art (LaRC) in October 2014. The Big Data Big Think innovation event, hosted by the museum’s CIO, meetings provide a forum where SMEs share Jane Alexander. The team also toured the GRC what they have been working on and collabo- Simulated Lunar Operations (SLOPE) Lab to see rate on solutions to NASA’s data challenges. the latest in rover technology. Visitors tour the Cleveland Museum of Art event. At center, using the interactive touch The meeting was opened by GRC Deputy Direc- Many thanks to the GRC CIO and staff for host- display, is NASA data scientist Dr. Brian tor Dr. Marla E. Pérez-Davis, followed by GRC ing this event! Thomas. Photo by Herb Schilling, GRC

Comprehensive Digital Transformation Initiative Ignites Data Science Adoption at NASA’s Langley Research Center By Jeremy Yagle, Technical Lead, LaRC Data Science Team; Corey Portalatin-Berrien, LaRC Data Science Team; and Ed McLarney, Strategic Lead, LaRC Data Science Team—Langley Research Center

Data science is the convergence of statistics, referenced above and created the Compre- Project success is dependent upon collabora- social sciences, and computer sciences. The hensive Digital Transformation (CDT) initiative. tion between LaRC’s OCIO Data Science team, discipline aims to discover the underlying value The primary goal of CDT is to catalyze the ap- mission organizations, and external partners in data, from identifying trends to analyzing plication of cutting-edge capabilities to enable from academia and industry. This convergence massive datasets for anomalies. Data science innovative concepts, reduce design cycle time, of skills provides a comprehensive view of the includes the subdomains of machine learning, improve affordability, and increase confidence. problem while encouraging collaboration, buy- big data analytics, and artificial intelligence (AI). in, and continued learning. Speech-to-text on smartphones, online shop- In 2014, LaRC’s Big Data Analytics and Machine ping recommendation systems, and smart- Intelligence Strategy and Roadmap outlined a Since there are few off-the-shelf solutions for watch fitness analysis are just a small sampling long-term vision for an AI-driven “Virtual Re- applying these techniques to physics-based of technologies that have rapidly become an in- search and Design Partner” to enable NASA datasets at NASA, our challenges must be re- tegral part of daily life. to achieve greater scientific discoveries. A key searched and solutions developed by enhanc- component was the creation of LaRC’s Office of ing current methods and providing regular out- Significant advances in data science methods the Chief Information Officer (OCIO) Data Sci- reach. The OCIO Data Science team is working have demonstrated the potential to transform ence team. Since its inception, the LaRC Data to build a robust data architecture and infra- the disciplines of science and engineering. Science team has made significant progress structure that will be used by multiple disci- Novel approaches, paired with rapidly evolv- with the implementation of more than two doz- plines to gain new insights into heterogeneous ing computational capabilities, create an envi- en pilot projects and use cases across multiple technical data. ronment where cross-fertilization and unique domains. Projects include the exploration of ap- collaborations achieve previously unattainable plications of machine learning on images, the The LaRC OCIO Data Science Team is an enthu- outcomes. analysis of sensor data for humans and materi- siastic member of the Agency big data commu- als, the development of computational models, nity and enjoys partnering with other Centers to Data Science at Langley and the analysis of mission support datasets. benefit the NASA mission. Four years ago, NASA’s Langley Research Cen- These projects have allowed the team to assess ter (LaRC) recognized the technology changes potential value for NASA researchers.

NASA OCIO IT Talk Jul - Sep 2018 9 NASA/GSFC … Home of Big Data By Myra Bambacus, Program Manager, Program Formulation & Development Office, GSFC POC for Big Data, and Chair of the National Science Foundation Spatiotemporal Center Industry Advisory Board, and Daniel Duffy, Chief of the Computational and Information Sciences and Technology Office, Goddard Space Flight Center

To the scientists and engineers at NA- that govern the Earth’s environment in Center (STC). Lead by George Mason SA’s Goddard Space Flight Center order to design the next generation sat- University with co-leads at University of (GSFC), it seems unnecessary to point ellites and remote sensing platforms. California, Santa Barbara, and Harvard out that Goddard is a big data hub for Starting from observation data, com- University, the STC supports Goddard Earth science, planetary, heliophysics, puter models (see Figure 2) simulate initiatives in cloud computing infra- astrophysics, and engineering data. complex processes in the Earth sys- structure, planetary defense mitigation Over the past 60 years, Goddard has tem and are used for data assimilation, gateway, and climate data analytics. lead and been a part of numerous mis- research weather forecasts, decadal sions studying the Earth, sun, planets, studies, retrospective analysis, and cli- Given the size and complexity of the and the universe, all of which have gen- mate studies. It is not unusual for appli- data, downloading data is no longer an erated big data. While missions come cations to generate petabytes of data. option. Co-locating data, either phys- and go, the data that NASA generates ically or virtually, with compute capa- lives on. Within the Earth Science Technology bilities is a requirement. By combining Office (ESTO), the Advanced Informa- large-scale data volumes with elastic In addition to generating volumes of tion Systems Technology (AIST) Pro- compute capabilities, researchers can data, major NASA programs are at the gram develops and advances Infor- scale from small to large problems ef- forefront of managing access to data mation Technologies (IT) needed for ficiently. in new and innovative ways. The Earth NASA Earth Science programs in the 5 Observing System Data and Informa- to 25-year timeframe, providing tools, NASA is only beginning to understand tion System (EOSDIS) program man- technologies, and virtual environments the applications of big data. Artificial ages NASA’s Earth science data from in which big data science is accom- Intelligence (AI) has created new areas remote sensing instruments, and be- plished. of research that require access to large yond. EOSDIS curates a multi-pet- disparate data sets to train models to abyte archive distributed across mul- NASA cannot do all of this alone and answer a variety of scientific questions. tiple data centers that provide search, is constantly looking for external part- Goddard is rapidly moving into the next order, data distribution, and analytics nerships to help accelerate our mis- phase of big data technologies using services. Future missions will require sion. One such partnership which is models and algorithms to harness the more data storage so EODIS is looking leveraging millions of dollars of invest- potential of AI. By both using existing to the cloud as a solution. ments across academia, industry, and technologies and NASA innovations, the government is the National Sci- Goddard is enabling new ways big data At the NASA Center for Climate Sim- ence Foundation’s (NSF’s) Industry– can change how people act, play, and ulation (NCCS), scientists are increas- University Cooperative Research Cen- thrive on Earth and among the stars. ing our understanding of the processes ters Program (I/UCRC) Spatiotemporal

Figure 1: Two programs containing 100’s of petabytes Figure 2: Goddard Earth Observing System (GEOS) global that represent the Big Data Challenges in NASA. atmospheric model. Image courtesy of the GMAO.

10 www.nasa.gov BSA Corner By Meredith Isaacs, Communications Specialist, NASA Headquarters

NASA began implementing its IT Busi- ness Services Assessment (BSA) de- cisions in March 2016. The Mission Support Council (MSC) approved IT transformation decisions for Roles and Office 365 Responsibilities, Governance, Comput- ing Services, Communications, Work- Coming to NASA stations, Collaboration, and IT Security. By Emily Townsend, EUSO Communications Lead, Marshall During this implementation period, Space Flight Center IT at NASA has journeyed from limit- Prior to IT BSA, some Communications ed investment insight, duplicative ser- services were transforming to enter- Office 365 will allow NASA employ- vices, and minimal stakeholder en- prise services. IT BSA efforts address ees to work in the cloud AND on the gagement. While still transforming in a transformation risk and reinvestments ground. The Agency’s migration to Of- number of ways, IT at NASA is identi- of savings. Center End User Services fice 365’s cloud-based service starts fiable by greater CIO insight and over- procurement was varied and did not al- this month with a limited pilot and will sight, more efficient operations, formal ways take advantage of enterprise ser- continue throughout the summer and assessments, and the implementation vices. Today, nearly all Centers procure into the Fall. Initially, NASA will imple- of new tools. Each IT BSA decision area at least 80% of their workstations from ment Online Email, Skype, and One- has made great progress and four have the enterprise contract and the pro- Drive (which includes personal storage completed implementation. gram continues further contract con- and file sync and share within NASA). solidation. In the meantime, Microsoft Office (i.e., For Roles and Responsibilities, there Word, Excel, Access, and PowerPoint) was limited insight into IT spending Like other IT services, many of NASA’s will continue to function as they do to- and an operating model that didn’t re- collaboration tools were duplicated at day. Other O365 features, including ad- flect all program areas. Now, there are each Center, raising costs. By defining ditional collaboration capabilities, roll six IT programs (Applications, Com- a Core Suite of Collaboration Tools to out in Spring 2019. munications, Computing Services, End meet most needs and filling the Core User Services, Information Manage- Suite with Cisco capabilities and Office All NASA email-enabled customers ment, and Cybersecurity Services) and 365 (coming Fall 2018), we will reduce will receive an Office 365 account. Be- greater CIO oversight and authority, fol- duplicative services and save Agen- cause securing our devices, networks, lowing Federal Information Technology cy resources. Cybersecurity teams ad- and data are top priorities for NASA, Acquisition Reform Act (FITARA) guide- dressed potential gaps in coverage by Office 365 will be configured as an -in lines. adopting a risk management frame- ternal service and will require an ap- work and deploying continuous moni- proved NASA device and PIV card or In Governance, boards were ineffec- toring tools. Agency Smart Badge to gain access. tive and IT lacked insight mechanisms. Customers will be migrated in a phased Now, governance boards are stream- Through IT BSA activities, NASA’s IT approach beginning late summer and lined and thriving, and the IT Capital has evolved into a state of greater CIO will receive targeted communication, Investment Review and Center Func- oversight, more enterprise services, instructions, and training close to their tional Review have been greater service consistency, in- migration date. implemented. Before IT sight-providing assessments, BSA and Data Cen- savings, and strategic re- Office 365 is a smart decision for NASA. ter Optimization Ini- source management. IT It provides the agency with modern, in- tiative efforts, there continues transformation novative, and collaborative technolo- were 79 data cen- journey through final IT gy and gives our customers increased ters. Now, there are BSA efforts and in initia- mailbox sizes, online storage capabil- 24 and NASA has tives to follow. ities with OneDrive, and larger atten- adopted a Cloud dance per Skype meeting. First strategy and imple- For more information, FAQs, O365 train- mented a ing videos, and a list of Center POCs, visit: data cen- https://inside.nasa.gov/office-365. ter architec- ture.

NASA OCIO IT Talk Jul - Sep 2018 11 Center Functional Review By Saurabh Baveja, CFR Lead, NASA Headquarters

Over the past few years, the Business Services Assessment ing ensure alignment and compliance with the requirements (BSA) efforts have spanned the Agency, including how infor- levied per the BSA, as well as from the strategic direction of mation technology (IT) at NASA works and how best to op- the OCIO. timize IT to execute projects that benefit and provide added value to our customers, providers, and the Agency. IT Program Areas: Crosscutting Areas: Applications Collaboration Based on this effort, the Mission Support Council (MSC) de- Communications Enterprise Architecture veloped a BSA IT Implementation Plan with the goal of creat- Computing Governance ing a Level 0 through Level 3 management structure to enable End User IT Contracts a strategic path forward while clearly defining responsibilities Information Management IT Investments and authorities for each level. As a result, the Implementa- IT Security Project Management tion Plan approach established six IT programs, with each being spearheaded by a Program Executive (at Level 2) and The next CFR is scheduled at Johnson Space Center (JSC) in their respective Program Offices operating at Level 3. The July 2018, and it will mark the midpoint milestone of this ac- plan further laid out key decisions in the area of Governance tivity. Per the Implementation Plan, these reviews will rotate including the requirement that the CIO shall conduct Center on a 3-year cycle. Past CFRs have been successfully held Functional Reviews (CFR). These reviews, per the Business at Langley Research Center (LaRC), Kennedy Space Center Services Steering Committee (BSSC) recommendation, were (KSC), Headquarters (HQ), and Ames Research Center (ARC). put in place to ensure Center compliance with Agency IT pol- icy and standards and serve as a “means to assess the effec- CFRs continue to serve a pivotal role in helping enable the tiveness, efficiency, and compliance of the IT work performed BSA Implementation and Governance Plan requirements. across Centers…through Agency Program Executives.” These CFRs, in partnership with the Center CIOs and Cen- ter leadership, have come to serve not only as a means to Over the course of FY 2017 and FY 2018, multiple successful assess and look closer at compliance with existing NASA CFRs have taken place to ensure alignment with these BSA- IT policies and standards, but also as an opportunity to im- prescribed guidelines. These reviews have come to span six prove communications. This partnership helps identify ad- IT programs and six crosscutting areas (see list) and have ditional enterprise adoption opportunities while increasing resulted in collaborative dialogue and improvements by the awareness and understanding of the Centers we (in IT) serve Centers, as well as programs and crosscutting areas, help- to help better meet current and future mission needs.

NASA Cloud Team Member Wins FedRAMP Award! By Penny Hubbard, Communications Specialist, Ames Research Center The Federal Risk and Authorization Management Program (FedRAMP) of the General Services Administration (GSA) an- nounced the winners of the inaugural FedRAMP Five Awards in Washington, DC, on June 13, 2018, during the Advanced Technology Academic Research Center’s (ATARC) Cloud and Data Center Summit. The FedRAMP Five is FedRAMP’s first- ever awards program and honors the agencies and people that are doing the most to shape program progress to en- able cloud providers to work with the Federal Government and provide approved and secure cloud services. NASA’s Computing Services Program Office (CSPO) is pleased to congratulate Steven Hunt, who was selected to receive Fe- dRAMP’s Large Agency Tech Lead award. This award is es- pecially significant because it is given to only a single indi- vidual across the entire Federal Government.

The full details are posted on the FedRAMP site here: https:// www.fedramp.gov/congratulations-to-the-2018-fedramp- five-award-winners/. Ames Research Center’s Steven Hunt, winner of the Congratulations, Steven! FedRAMP Five Large Agency Tech Lead award.

12 www.nasa.gov Cleveland Data Meetup By John Sprague, Deputy Associate CIO for Technology, Data & Innovation Division, NASA Headquarters

Data meetups are held all over the United States and the extravehicular activity (EVA), as well as unique challenges world. Cleveland, OH, holds some of the largest meetups from astronaut data. in the country. Attendees to these monthly and quarterly meetups are from the public and private sectors from the We discussed new space research, like the romaine lettuce surrounding area, and Cleveland is home to a number of grown and eaten by astronauts (including Scott Kelly) on the businesses with a focus on data management and analyt- International Space Station, as well as the anticipated wave ics, such as Progressive Insurance, Key Bank, Pandata, of data expected from the slew of missions scheduled over and IBM Watson Health. While in town for NASA’s Big Data the next year. Meetup participants were also interested in Big Think conference at Glenn Research Center (GRC), a NASA’s Distributed Active Archive Centers (DAACs), scat- number of NASA personnel were able to attend Cleveland’s tered across the United States and storing large amounts Big Data Meetup; I served as the keynote speaker for this of data. It was a great, free flow of information as we fielded particular meetup. numerous questions and held a dialogue. IT Talk

IT Talk July - September 2018 Volume 8 • Issue 3 Topics ran the gamut from NASA’s Pleiades Supercomput- Cleveland’s data meetup was an excellent opportunity for er to the Lunar Laser Communication Demonstration and NASA to talk about its data management challenges, initia- Quantum Artificial Intelligence Laboratory located at Ames tives, and accomplishments. After all of the discussions, I was Research Center (ARC). I shared use cases like the Hubble reminded of one of my favorite sayings: Data becomes Infor- Space Telescope, the James Webb Space Telescope, and mation, becomes Knowledge, which becomes Wisdom.

NASA’s John Sprague presents at the meetup.

National Aeronautics and Space Administration AA Office of the Chief Information Officer 300 E Street, SW Washington, DC 20546 www.nasa.gov