PDL Paper wins ACM SoCC Test of Time Award!
The ACM SoCC 2021 test of time award goes to "Heterogeneity and Dynamicity of Clouds at Scale: Google Trace Analysis"! Congratulations to the authors Charles Reiss, Alexey Tumanov, Gregory R. Ganger, Randy H. Katz, and Michael A. Kozuch for their impactful work. The paper was first introduced at the 3rd ACM Symposium on Cloud Computing, held from October 14th-17th, 2012, in San Jose, CA. The paper helps its readers to better understand the challenges in developing effective cloudbased resource schedulers, analyzing the first publicly available trace data from a sizable multi-purpose cluster.
Mochi Project Wins 2021 R&D 100 Award
Mochi is a state-of-the-art open source tool for rapid development of customized data services supporting high-performance computing, big data, and large-scale learning across many scientific fields. Mochi allows developing these distributed services with a relatively small amount of new code while still enabling very high performance. Mochi was co-developed by researchers from Argonne National Laboratory, Carnegie Mellon University (George Amvrosiadis, Chuck Cranor, Qing Zheng), and Los Alamos National Laboratory. Established in 1963, the R&D 100 Awards is the only S&T (science and technology) awards competition that recognizes new commercial products, technologies and materials for their technological significance that are available for sale or license.
Sara McAllister and Colleagues Win Best Paper at SOSP'21!
Congratulations to Sara McAllister, Benjamin Berg, Julian Tutuncu-Macias, Juncheng Yang, Sathya Gunasekar, Jimmy Lu, Daniel Berger, Nathan Beckmann, and Gregory R. Ganger on receiving the Best Paper Award at SOSP'21, which was held virtually in Germany. The paper "Kangaroo: Caching Billions of Tiny Objects on Flash" introduces Kangaroo, a new flash-cache design that optimizes both DRAM usage and flash writes to maximize cache performance while minimizing cost.
PDL Researchers Win Best Paper at OSDI'21!
Congratulations to Aurick Qiao, Sang Keun Choe, Suhas Jayaram Subramanya, Willie Neiswanger, Qirong Ho, Hao Zhang, Gregory R. Ganger, and Eric P. Xing, on being awarded one of three Best Paper Awards at OSDI'21. The paper "Pollux: Co-adaptive Cluster Scheduling for Goodput-Optimized Deep Learning" discusses improved scheduling performance in deep learning (DL) clusters by adaptively co-optimizing inter-dependent factors both at the per-job level and at the cluster-wide level.
PDL Alum, Ted Wong, Appears in Times Square!
We are pleased to share that PDL Alum, Ted Wong, is now a celebrity! Ted, whose company of employment, 23andMe, went public today had their employees appear on the big screen in Times Square! Ted says he is lucky to be at 23andMe and excited that he gets to work with a lot of talented and fun people.
Christos Faloutsos promoted to University Professor
Congratulations to Christos, who has been elevated to the rank of University Professor, the highest distinction a faculty member can achieve at CMU.
Huaicheng Li Recognized as Distinguished Reviewer
Congratulations to Huaicheng, who has been recognized as a Distinguised Reviewer for his service to Systor 2021, the 14th ACM International System and Storage Conference.
CMU Research Forms Basis for Automatic Database Tuning Service: SCS Alums, Faculty Found OtterTune
A School of Computer Science grad student's (Dana Van Aken) dissertation has been transformed into a service aimed at improving the databases that power popular websites. OtterTune, a play on the once ubiquitous Auto-Tune, uses machine learning to automatically optimize databases, improving performance and efficiency and potentially saving companies time and money. Users could see faster loading times and improved services with a database humming along in the background.
"There are hundreds of settings to consider when optimizing a database, too many for humans to properly tune," said Dana Van Aken, one of the company's co-founders. "OtterTune takes human trial and error out of the mix."
The company is based on Van Aken's dissertation. She founded the company with her advisor, Andy Pavlo, an associate professor in the Computer Science Department; and Bohan Zhang, who earned his master of computational data science from CMU and worked with Van Aken and Pavlo as a research assistant.
The company announced a commercial version of its service Wednesday and $2.5 million in seed funding, led by venture capital firm Accel. OtterTune is based in Pittsburgh and full of Carnegie Mellon talent. Seven of its 12 employees are either CMU faculty or alumni.
"Database management systems now exceed the administrator's ability to optimize them," said Pavlo, the CEO of OtterTune. "We've put years of research into solving this problem, which we know will lead to significant increases in efficiencies and cost savings for customers."
Don't let the company's playful name or DJing otters wearing headphones fool you. It is serious about tuning a database for optimal performance. In case studies, the OtterTune improved efficiency by 33% to 50%, cut one company's costs in half, and saved another tens of thousands of dollars. "It turns out, otters are actually vicious animals," Pavlo said.
-- Aaron Aupperlee, CMU SCS News Wednesday, May 12, 2021
Huanchen Zhang Wins Jim Gray Dissertation Award!
Congratulations to Huanchen Zhang, who has been awarded this year’s ACM SIGMOD Jim Gray Dissertation Award. The award recognizes excellent research by doctoral candidates in the database field. Huanchen's dissertation studied “Memory-Efficient Search Trees for Database Management Systems” and addressed the challenge of building compact yet fast in-memory search trees to allow more efficient use of memory in data processing systems.
Since graduation Huanchen has spent time at Snowflake as a Postdoctoral Research Fellow and has now joined Tsinghua University as an Assistant Professor. While at CMU, Huanchen was advised by David Andersen and Andy Pavlo.
Nathan Beckmann Awarded Google Research Scholar Grant
Nathan Beckmann, CSD and ECE has received a grant through Google's inaugural Research Scholar Program, which will support his research on "Making Data Access Faster and Cheaper with Smarter Flash Caches". The Research Scholar Program aims to support early-career professors who are pursuing research in fields relevant to Google.
-- info from SCS Wednesday Wire and research.google/
Juncheng Yang and Rashmi Vinayak Receive NDSI'21 Community Award and Best Paper!
Congratulations to Juncheng and Rashmi, who received theNSDI 2021 Community Award for their paper "Segcache: A Memory-efficient and Scalable In-memory Key-value Cache for Small Objects." The conference, held virtually this year, presents the award for the best paper whose code and/or data set is made publicly available by the final papers deadline. The paper also won the overall Best Paper Award! In collaboration with Twitter, Juncheng and Rashmi are building the next generation in-memory caching systems. Segcache enables high memory efficiency, high throughput, and excellent scalability, and demonstrates several important optimization techniques for large-scale web services.
Sara McAllister wins 5-year graduate fellowship
Congratulations to Sara, as she receives a 5-year graduate fellowship under the NSF Graduate Research Fellowship Program. The NSF GRFP recognizes and supports outstanding graduate students in NSF-supported STEM disciplines who are pursuing research-based master’s and doctoral degrees at accredited US institutions. The five-year fellowship includes three years of financial support including an annual stipend and a cost of education allowance. Sara will be pursuing her Ph.D. in Computer Science in the field of Computer Systems and Embedded Systems.
-- with info from nsfgrfp.org
Two PDL PhD Students Win Apple Fellowship in Integrated Systems
Congratulations to Minh S.Q. Truong and Mohammad Bakhshalipour, who have both received an Apple PhD Fellowship in Integrated Systems for the 2021-2023 academic years! The award includes a stipend and internships with Apple during the summers covered by the award.
Lorrie Cranor named AAAS Fellow
Lorrie Cranor, the director of CyLab and a professor in the Institute for Software Research and the department of Engineering and Public Policy, has been named a Fellow of the American Association for the Advancement of Science (AAAS). As part of the section on information, computing, and communication, Cranor was elected as an AAAS Fellow for her contributions to usable privacy and security research, policy, and education.
Election as an AAAS Fellow is an honor bestowed upon AAAS members by their peers and has been a tradition since 1874.
-- info from engineering.cmu.edu/news by Daniel Tkacik
More PDL News here.