Delta FS Project Takes Home R&D 100 Award
The R&D World Magazine announced its 100 winners for 2019 on October 29. "These 100 winning products and technologies are the disruptors that will change industries and make the world a better place in the coming years," said Paul J. Heney, Vice President, Editorial Director for R&D World.
In the IT/Electrical category, the winner is the CMU/Los Alamos National Laboratory collaborative project "DeltaFS—Rapidly Searching Big Data." Congratulations to its may contributors, including
Brad Settlemyer, Scientist, Los Alamos National Laboratory; George Amvrosiadis, Research Professor
Carnegie Mellon University; Gary Grider, HPC Division Director, Los Alamos National Laboratory; Qing Zheng, Research Assistant, Carnegie Mellon University; Greg Ganger, Jatras Professor, Carnegie Mellon University; Charles Cranor, Systems Scientist, Carnegie Mellon University; and Garth Gibson, Professor, Carnegie Mellon University.
--RDWorld Online, October 29, 2019
PDL Alum Joy Arulraj Receives SIGMOD Jim Gray Doctoral Dissertation Award
Joy Arulraj is an assistant Professor of Computer Science at Georgia Institute of Technology. He received his Ph.D. from Carnegie Mellon University in 2018, advised by Andy Pavlo. His doctoral research focused on the design and implementation of non-volatile memory database management systems. This work was conducted in collaboration with the Intel Science and Technology Center for Big Data, Microsoft Research, and Samsung Research.
-- sigmod.org news
Lorrie Cranor Made Member of ACM's New Technology Policy Council
Lorrie Cranor, the FORE Systems Professor of Computer Science and Engineering and Public Policy, has joined 11 other distinguished computer scientists on the Technology Policy Council, newly formed by the Association for Computing Machinery (ACM). The ACM, the world’s largest association of computing professionals, created the council to coordinate the agenda for its policy activities around the globe. It also will serve as the ACM’s contact point for its interaction with government organizations, the computing community and the public in matters of public policy related to information technology and computing. Cranor is the director and Bosch Distinguished Professor in Security and Privacy Technologies at the CyLab Security and Privacy Institute. She also directs the CyLab Usable Privacy and Security Laboratory and co-directs the privacy engineering master’s program. In 2016, she served as the chief technologist of the Federal Trade Commission.
-- The Piper, September 24, 2019
Abutalib Aghayev Awarded Hima and Jive Graduate Fellowship
Congraulations to Abutalib on receiving the Hima and Jive Fellowship this year! An anonymous donor established the Hima and Jive Fellowship in Computer Science for International Students in 2012 to support one third-year graduate student annually in the Computer Science Department who has a permanent residence outside the United States, regardless of their national origin. This fellowship is to encourage students to overcome challenges and to have fun doing it. The fellowship is given to one international student in the School of Computer Science annually.
Beckmann Earns NSF Early CAREER Award
Nathan Beckmann, an assistant professor in the Computer Science Department, has received a Faculty Early Career Development Award, the NSF's most prestigious award for young faculty members.
Nathan Beckmann, an assistant professor in the Computer Science Department, has received a five-year, roughly $500,000 Faculty Early Career Development (CAREER) Award, the National Science Foundation's most prestigious award for young faculty members.
Beckmann's research interests include computer architecture and performance modeling. The NSF grant will support his work crafting and evaluating a new computer system design that makes accessing data faster and cheaper. Beckmann said more energy efficiency is needed to sustain growth in computing power for machine learning, social networking and robotics.
Applications currently have no control over how data is managed because memory hierarchy is fixed in hardware and hidden from software, resulting in unnecessary data movement. Beckmann's project will develop a new hardware-software co-design, wherein the operating system and hardware will collaboratively schedule tasks and data to improve efficiency.
Beckmann will involve high school, undergraduate and graduate students in research. He will also organize research workshops for undergraduate women and a summer internship program for underrepresented minorities.
Beckman earned his master's degree and Ph.D. from the Massachusetts Institute of Technology, where he spent one year post-doc in the Computer Science and Artificial Intelligence Laboratory.
-- SCS News, Virginia Alvino Young, September 12, 2019
George Amvrosiadis Reports on the Future of Storage
George Amvrosiadis led 33 scientists from academia, industry, and federal agencies in the compilation of a report on future storage research for the National Science Foundation (NSF). Their Data Storage Research Vision 2025 recommends effort in four key areas for innovative research and education: enhancing cloud and edge computing I/O infrastructures; designing storage for emerging AI applications; rethinking the storage systems abstractions in service of for new and innovative applications; and redesigning storage systems for emerging hardware.
Schwedock Receives NSF Graduate Research Fellowship
Brian Schwedock, an electrical and computer engineering Ph.D. student, has received the prestigious National Science Foundation (NSF) Graduate Research Fellowship for his work in computer architecture and computer systems with a focus on caching.
Schwedock’s current project improves the performance and energy efficiency of chip-multiprocessors in data centers. Data centers waste significant amounts of hardware, energy, and capital by isolating applications with different priorities, specifically latency-critical applications and batch applications.
“My project proposes an operating system runtime which reduces this waste by intelligently sharing hardware caches among these different applications,” says Schwedock. “Our results show major improvements in performance and energy efficiency for low priority batch applications while still meeting strict deadlines required by high priority latency-critical applications.”
The NSF Graduate Research Fellowship Program recognizes and supports outstanding graduate students in NSF-supported science, technology, engineering, and mathematics disciplines who are pursuing research-based Master's and doctoral degrees at accredited United States institutions.
Schwedock is advised by Nathan Beckmann, assistant professor in the Computer Science Department.
Congratulations are also due to Giulio Zhou, who received an honorable mention for the NSF Graduate Research Fellowship Program this year.
-- ECE News and Events - April 18, 2019
Joy Arulraj Wins SIGMOD Jim Gray Dissertation Award
The Carnegie Mellon Database Group is pleased to announce that CS alumnus Prof. Joy Arulraj (now faculty at the Georgia Institute of Technology) has won the 2019 ACM SIGMOD Jim Gray Dissertation Award. This honor is conferred for the best dissertation in the field of databases for the previous year. Joy’s thesis, entitled “The Design and Implementation of a Non-Volatile Memory Database Management Systems”, is based on his work exploring new DBMS architectures for NVM. This was work done in collaboration with Intel Labs as part of the Intel Science & Technology Center for Big Data. Joy’s research findings were also was published in 2019 as the book “Non-Volatile Memory Database Management Systems” from Morgan & Claypool.
-- Carnegie Mellon University Database Group News, April 28, 2019
Lorrie Cranor Recieves University of Washington 2019 Alumni Achievement Award
Washington University in St. Louis and the James McKelvey School of Engineering has honored Lorrie Cranor with its 2019 Alumni Achievement Award, praising her passion for teaching and for creating in her wide ranging career an interdisciplinary program leveraging science, humanities and the arts. She is an accomplished scholar and CMU President Farnam Jahanian describes her research as "vigorous, multifaceted and highly relevant to society." Lorrie is a Professor of Computer Science and Engineering and Public Policy at CMU and focuses on usable privacy and security, helping to make us safe and secure.
As a graduate student at Washington University, she was a Woodward scholar, editor for the Student Life newspaper, founder of the Student Association for Engineering graduate students, and the graduate student representative to the board of trustees, among other roles. She is still involved with WashU as member of the Dept. of Engineering and Computer Science external advisory board and volunteers with the parent and alumni admission program. She calls her parents role models for her efforts to excel in the many facets of her academic career.
Lorrie Cranor Named Andrew Carnegie Fellow
Carnegie Mellon University faculty member Lorrie Cranor has been named to the 2019 Class of Andrew Carnegie Fellows by the Carnegie Corporation of New York, a philanthropic foundation that has supported the advancement of education and knowledge for more than a century. She is one of 32 distinguished scholars and writers selected from nearly 300 nominations.
"Andrew Carnegie believed in education and understood its influence on the progress of society and mankind. The Andrew Carnegie Fellows Program is an integral part of carrying out the mission he set for our organization," said Vartan Gregorian, president of Carnegie Corporation of New York. "We salute this year's class and all of the applicants for demonstrating the vitality of American higher education and scholarship."
Cranor is director and Bosch Distinguished Professor in Security and Privacy Technologies of the CyLab Privacy and Security Institute, is the FORE Systems Professor of Computer Science and of Engineering and Public Policy and directs the CyLab Usable Privacy and Security (CUPS) Laboratory. Furthermore, she co-directs Carnegie Mellon's Privacy Engineering master's program and served as chief technologist at the Federal Trade Commission in 2016.
Having authored over 150 papers on privacy and security, Cranor's work has championed usability as a key element to making modern technology more safe, secure, and privacy-aware. Her seminal book, "Security and Usability," is widely regarded as a foundational work in the field. As well, the Symposium On Usable Privacy and Security (SOUPS), which Cranor founded, is considered the preeminent conference on the topic.
"This recognition by the Carnegie Corporation is very meaningful, not only on a personal level, but also in terms of the work ahead of us in privacy and security," Cranor said. "Our lack of attention to the ways that humans interact with security and privacy tools has led to security vulnerabilities and public policies that fail to protect security and privacy. This gift will enable me to pursue further work on developing, validating, and documenting empirical methods for the study of security, privacy, and human behavior so that these methods might be more readily applied in evidence-based decision making by policy makers as well as within organizations."
-- abridged from CMU News, by Joshua Quicksall & Caitlin Kizielewicz, April 23, 2019.
Lorrie Cranor Elected to CRA Board of Directors
Lorrie Faith Cranor has been elected to the Computing Research Association board of directors. Cranor is the director and Bosch Distinguished Professor in Security and Privacy Technologies of CyLab and FORE Systems Professor of Computer Science and of Engineering and Public Policy at Carnegie Mellon University. She also directs the CyLab Usable Privacy and Security Laboratory (CUPS) and co-directs the MSIT-Privacy Engineering Master’s program. In 2016, she served as chief technologist at the U.S. Federal Trade Commission. She also is a co-founder of Wombat Security Technologies, Inc, a security awareness training company.
-- The Piper, CMU Community News, March 21, 2019
Best Paper at NSDI'19!
Congratulations to Anuj Kalia, Michael Kaminsky (intel), and David Andersen for receiving the Best Paper Award for their work on "Datacenter RPCs can be General and Fast" at NSDI'19 (Networked Systems Design and Implementation). The paper addresses datacenter neworking efficiencies for high performance and shows that specialized distributed systems designed specifically for niche technologies such as RDMA, lossless networks, FPGAs, and programmable switches testifies to this belief are not necessary.
Nathan Beckmann and Rashmi Vinayak Receive Google Faculty Research Awards
Congratulations to Nathan Beckmann and Rashmi Vinayak, both Assistant Professors of Computer Science, on being awarded Google Faculty Research Awards. Rashmi's reserach focuses on the broad area of computer and networked systems with current attention on reliability, availability, scalability, and performance challenges in data storage and caching systems, in systems for machine learning and in live video streaming. Nathan is interested in computer systems, computer architecture, and performance modeling. His current projects cover hardware, software, and theory and he is currently working on how to build intelligent edge devices, make parallel systems more efficient (particularly by making it less expensive to access data), and use theory to address the huge practical problems still posed by caches.
The Google Faculty Research Awards provide unrestricted gifts as support for research at institutions around the world. The program is focused on funding world-class technical research in Computer Science, Engineering, and related fields.
Lorrie Cranor Receives the Bosch Chair
CyLab Director Lorrie Cranor has received the Bosch Distinguished Professorship in Security and Privacy Technologies, enabling her to lead a new era of security and privacy research at Carnegie Mellon. The Bosch Chair provides funding support for groundbreaking research addressing important issues related to the security of our connected environment and the privacy of personal data. At the same time, the Bosch Chair affords recognition for the work and career achievements of the CyLab director. “With Lorrie and CyLab as our partner, we at Bosch are looking forward to the opportunity to help shape the future of security and privacy research, together with the world’s leading institution on the topic," said Sylvia Vogt, president of the Carnegie Bosch Institute. In addition to serving as the director of CyLab, Cranor is a professor in the Institute for Software Research and in the Department of Engineering and Public Policy, and she serves as co-director of Carnegie Mellon’s Privacy Engineering master’s degree program.
-- The Piper, CMU Community News
Lorrie Faith Cranor Named New Director of Carnegie Mellon University's CyLab
Lorrie Faith Cranor has been named the next director of CyLab, Carnegie Mellon University's security and privacy institute, effective January 15. As Director, Cranor will be assuming the chair as the Bosch Distinguished Professor in Security and Privacy Technologies. CyLab, founded in 2003, brings together security and privacy experts from all schools across Carnegie Mellon with the vision of creating a world in which technology can be trusted.
"I'm honored and thrilled to serve as CyLab's next director," Cranor said. "I look forward to supporting CyLab's ongoing success and bolstering research aimed at making our increasingly digital world safe and trustworthy."
Cranor is the FORE Systems Professor of Computer Science and of Engineering and Public Policy, and directs the CyLab Usable Privacy and Security (CUPS) Laboratory. She is a co-director of Carnegie Mellon's Privacy Engineering master's program, and served as Chief Technologist at the Federal Trade Commission (FTC) in 2016.
An internal committee conducted a rigorous international search for director candidates. Cranor was selected for her leadership in the field and for her vision of the next phase of CyLab's growth.
"Lorrie's extensive leadership experience and background, as well as her recent government experience as the FTC's Chief Technologist, make her an exceptional choice as CyLab's new director," said Jon Cagan, interim dean of Carnegie Mellon's College of Engineering.
Having played a key role in building the usable privacy and security research community, Cranor co-edited the seminal book Security and Usability and founded the Symposium On Usable Privacy and Security (SOUPS). She is a co-founder of Wombat Security Technologies, Inc, a security awareness training company.
Cranor has authored over 150 research papers on online privacy, usable privacy and security, and other topics. Her current research projects include password usability and security, privacy for the Internet of Things, and development of meaningful and usable privacy notices and consent experiences.
Before joining the Carnegie Mellon faculty, Cranor received her doctorate degree from Washington University in St. Louis and was a member of the secure systems research group at AT&T Labs-Research. She is a Fellow of both the Association for Computing Machinery (ACM) and the Institute of Electrical and Electronics Engineers (IEEE), and she is a member of the ACM CHI Academy.
Cranor's appointment follows that of Douglas Sicker, Head of Carnegie Mellon's Engineering and Public Policy department, who has served as CyLab's Interim Director since September 1, 2017. Sicker stepped in after the previous director, Electrical and Computer Engineering professor David Brumley, took a leave of absence to help grow his startup company, ForAllSecure.
-- Daniel Tkacik, CyLab Security and Privacy Institute News, Jan 14, 2019
Mor Harchol-Balter Receives Joel and Rut Spira Excellence in Teaching Award
Congratulations to Mor Harchol-Balter on being recognized for her teaching excellence. The Joel and Ruth Spira Excellence in Teaching Award is presented to an ECE faculty member based on major accomplishments and outstanding qualities and strengths in teaching. Established with an endowment by the Lutron Foundation, this award recognizes individuals who excel in the classroom by helping students learn, understand and apply the fundamentals of engineering.
Joshi Optimizes Computing Systems for IBM's Watson
Machine learning has grown dramatically in engineering and computer science in recent years with the explosion of interest in artificial intelligence. In machine learning, humans — engineers and computer scientists — feed large data sets into a neural network model to train the model to learn from data and eventually identify and analyze patterns and make decisions.
Carnegie Mellon University's Gauri Joshi is researching the analysis and optimization of computing systems. Joshi, assistant professor of Electrical and Computer Engineering (ECE), was has been named a recipient of a 2018 IBM Faculty Award for her research in distributed machine learning. Faculty Award recipients are nominated by IBM employees in recognition of a specific project that is of significant interest to the company and receive a cash award in support of the selected project.
Joshi's research is about distributing deep learning training algorithms. The datasets used to train neural network models are massive in size, so a single machine is not sufficient to handle the amount of data and the computing required to analyze the data. Therefore, datasets and computations are typically divided across multiple computing nodes (i.e. computers, machines, or servers), with each node responsible for one part of the data set.
In a distributed machine learning system with data sets divided across nodes, researchers use an algorithm called stochastic gradient descent (SGD), which is at the center of Joshi's research. The algorithm is distributed across the nodes and helps achieve the lowest possible error in the data. It requires exact synchronization, which can lead to delays.
"My work is about trying to strike the best balance between the error and the delay in distributed SGD algorithms," Joshi said. "In particular, this framework fits well with the IBM Watson machine learning platform; I will be working with the IBM Watson Machine Learning vision; I will be working with the IBM Research AI team."
In every iteration of the SGD, a central server is required to communicate with all of the nodes. If any of the nodes slow down, then the entire network slows down to wait for that node, which can significantly reduce the overall speed of the computation. Efficiency and speed of computation are the two main things Joshi aims to improve, both without risking the accuracy of the network.
"When you have a distributed system, communication and synchronization delays in the system always affect the proponents of the algorithm. I'm trying to design robust algorithms that work well on unreliable computing nodes," she said.
Prior to joining Carnegie Mellon's College of Engineering in fall 2017, Joshi was a research staff member at IBM's Thomas J. Watson Research Center. Because of her past experience, she was aware of the specific research projects that are relevant to the company's interests.
The funding provided by the Faculty Award will be used to support Joshi's students, who are working on the theoretical analysis for this project. In the future, she hopes to release an open source implementation of the new algorithm they have developed. Joshi plans to work with IBM to make this method available to anybody who wants to train their own machine learning algorithms using distributed SGD.
-- Marika Yang, Carnegie Mellon University News, January 9, 2019.
Mor Harchol-Balter made an IEEE Fellow
Mor Harchol-Balter has been elevated to fellow status in the Institute of Electrical and Electronics Engineers (IEEE), the world's largest technical professional organization. Fellow status is a distinction reserved for select members who have demonstrated extraordinary accomplishments in an IEEE field of interest. Mor, a professor in CSD since 1999, was cited "for contributions to performance analysis and design of computer systems." Her work on designing new resource-allocation policies includes load-balancing policies, power-management policies and scheduling policies for distributed systems. She is heavily involved in the SIGMETRICS/PERFORMANCE research community and is the author of a popular textbook, "Performance Analysis and Design of Computer Systems."
-- The Piper, CMU Community News, Dec. 12, 2018
PDL Team Designing Record-breaking Supercomputing File System Framework at Los Alamos National Lab
Trinity occupies a footprint the size of an entire floor of most office buildings, but its silently toiling workers are not flesh and blood. Trinity is a supercomputer at Los Alamos National Laboratory in New Mexico, made up of row upon row of CPUs stacked from the white-tiled floor to the fluorescent ceiling.
The machine is responsible for helping to maintain the United States’ nuclear stockpile, but it is also a valuable tool for researchers from a broad range of fields. The supercomputer can run huge simulations, modeling some of the most complex phenomena known to science.
However continued advances in computing power have raised new issues for researchers.
“If you find a way to double the number of CPUs that you have,” says George Amvrosiadis “you still have a problem of building software that will scale to use them efficiently.” He’s an assistant research professor in Carnegie Mellon’s Parallel Data Lab.
Amvrosiadis was part of a team, including Professors Garth Gibson, and Greg Ganger, Systems Scientist Chuck Cranor, and Ph.D. student Qing Zheng. The team recently lent a hand to a cosmologist from Los Alamos struggling to simulate complex plasma phenomena. The problem wasn’t that Trinity lacked the power to run the simulations, but rather, that it was unable to create and store the massive amounts of data quickly and efficiently. That’s where Amvrosiadis and the DeltaFS team came in.
DeltaFS is a file system designed to alleviate the significant burden placed on supercomputers by data-intensive simulations like the cosmologist’s plasma simulation. When it comes to supercomputing, efficiency is the name of the game. If a task can’t be completed within the amount of time allotted, then the simulation will go incomplete, and precious time will have been wasted. With researchers vying for limited computing resources, any time wasted is a major loss.
DeltaFS was able to streamline the plasma simulation, bringing what had once been too resource-demanding a task within the supercomputer’s capabilities by tweaking a couple parts of how Trinity processed and moved the data.
First, DeltaFS changed the size and quantity of files the simulation program created. Rather than taking large snapshots encompassing every particle in the simulation—which numbered more than a trillion—at once, DeltaFS created a much smaller file for each individual particle. This made it much easier for the scientists to track the activity of individual particles.
Through DeltaFS, Trinity was able to create a record-breaking trillion files in just two minutes. Additionally, DeltaFS was able to take advantage of the roughly 10% of simulation time that is usually spent storing the data created, during which Trinity’s CPUs are sitting idle. The system tagged data as it flowed to storage and created searchable indices that eliminated hours of time that scientists would have had to spend combing through data manually. This allowed the scientists to retrieve the information they needed 1,000–5,000 times faster than prior methods.
The team could not have been more thrilled with the success of DeltaFS’ first real-world test run and are already looking ahead to the future. “We're looking to get it into production and have the cosmologist who originally contacted us use it in his latest experiment,” says Amvrosiadis. “To me that's more of a success story than anything else. Often a lot of the work ends with just publishing a paper and then you're done; that’s just anticlimactic.”
But he and the rest of the team aren’t just looking to limit their efforts to cosmological simulations. They’re currently looking at ways to expand DeltaFS for use with everything from earthquake simulations to crystallography. With countries across the globe striving to create machines that can compute at the exascale, meaning 1018 calculations per second, there’s a growing need to streamline these demanding processes wherever possible.
The trick to finding a one-size-fits-all (or at least most) replacement for the current purpose-built systems in use, is designing the file system to be flexible enough for scientists and researchers to tailor it to their own specific needs.
“What researchers end up doing is stitching a solution together that is customized to exactly what they need, which takes a lot of developer hours,” says Amvrosiadis. “As soon as something changes they have to sit back down to the drawing board and start from scratch and redesign all their code.”
Amvrosiadis and the team have already demonstrated a couple of ways that efficiency can be improved, such as indexing or altering file size and quantity. Now they’re looking into further ways to take advantage of potential inefficiencies, like using in-process analysis to eliminate unneeded data before it ever reaches storage or compressing information in preparation for transfer to other labs.
Solutions like these center around repurposing CPU downtime to perform tasks that will contribute back into the information pipeline and creating smarter ways to organize and store data, increasing overall efficiency. The idea is to let the expert scientists identify the areas where they have room for improvement or untapped resources, and to take advantage of the toolkit and versatile framework DeltaFS can provide.
As the world moves toward exascale computing, the pace that software development must maintain to keep pace with hardware improvements will only increase. Amvrosiadis even hopes that one day more advanced AI techniques could be incorporated to do much of the observational work performed by scientists, cutting down on observation time and freeing them to focus on analysis and study. But for him and the rest of the DeltaFS team, all of that starts with finding little solutions to improve huge processes.
“I don’t know if there’s one framework to rule them all yet—but that’s the goal.”
-- Dan Carroll, CMU Engineering News, December 1, 2018.
More PDL News here.