Profile cover photo
Profile photo
Intel S&TC for Big Data
412 followers -
The Intel Science & Technology Center for Big Data (located at MIT CSAIL)
The Intel Science & Technology Center for Big Data (located at MIT CSAIL)

412 followers
About
Intel S&TC for Big Data's posts

Post has attachment
Intel S&TC Researcher Stavros Papadopoulos of +Intel and +CSAIL MIT presents TileDB, our novel efficient data management system for scientific data. TileDB is being used in the production pipeline at the +Broad Institute, an international leader in genomics.

TileDB and its adapation, GenomicsDB, are available in open source: http://istc-bigdata.org/index.php/tiledb-and-genomicsdb-now-available-in-open-source/


Post has shared content
Intel S&TC researcher and data visualization expert +Jeffrey Heer of the University of Washington is speaking at +SXSW March 13 at 9:30 AM. Learn more about his work this post on the ISTC blog http://istc-bigdata.org/index.php/higher-level-tools-for-interactive-data-visualization/
On March 13, Jeff Heer will host one of ACM’s sponsored sessions at #SXSW. Heer is an Associate Professor at the University of Washington, and the Co-Founder & Chief Experience Officer, at Trifacta Inc. Entitled “Interactive Data Analysis: Visualization & Beyond,” Heer’s session will consider how to accelerate people's exploration and understanding of data by crafting a careful balance of interactive and automated methods. We will examine how to combine concepts from data visualization, machine learning, and computer systems to design novel interactive analysis tools, covering both interactive demos and principles from academic research. #ACMatSXSW

Read more about ACM sessions and events at #SXSWInteractive: ow.ly/ZGbt308OToa
Photo

Post has attachment
If you're attending Spark Summit East 2017 this week in Boston (Feb 7-9) and you work with machine learning, you might want to attend the talk by +Manasi Vartak of +CSAIL MIT: "ModelDB: A System to Manage Machine Learning Models" 2/8 3:20 PM Rm. 312. You can learn more about ModelDB here in her blog post: http://istc-bigdata.org/index.php/modeldb-a-system-for-managing-machine-learning-models/


Post has attachment
Is the data behind that news story really legit? Maybe, but increasingly maybe not, particularly as new statistical analysis tools put visual, point-and-click data exploration into the hands of many more people, including novice users. In a recent paper, Intel S&TC researchers at +Brown University explore how such tools lead to false discoveries - and the possible downstream fallout. They've developed QUDE, a new system for automatically controlling the various risk factors during the data exploration process. Read the full paper here: http://cidrdb.org/cidr2017/papers/p56-binnig-cidr17.pdf and this excellent analysis of the paper by "The Morning Paper" blog.

Paper authors are Carsten Binnig, Lorenzo De Stefani, +Tim Kraska Eli Upfal, Emanuel Zgraggen and Zheguang Zhao.

Post has attachment
New on the blog: Intel S&TC PIs, researchers and their students presented their latest research work at the 9th Annual Northeast Database Day at MIT. Good snapshot of some of the ISTC work happening at +CSAIL MIT +Brown University and +Carnegie Mellon School of Computer Science (SCS).

North East Database Day 2017 showcased work by some of the top university research programs and corporate/industrial research labs operating in the Northeast.

Post has attachment
Program posted for 2017 Northeast Database Day, which takes place Friday, 1/27 at +CSAIL MIT in Cambridge, Mass. Cutting-edge research on database technology, presented by researchers from +Massachusetts Institute of Technology (MIT) +Harvard University, +Carnegie Mellon University +Brown University, MIT Lincoln Laboratory, +Intel, University of Massachusetts Amherst, +Facebook, HPE Vertica, MITRE and more. Keynotes:

John Leonard (MIT Department of Mechanical Engineering and MIT CSAIL)
"Self-Driving Vehicles, SLAM, and Databases"

+David J. DeWitt (MIT CSAIL), Willis Lang (Microsoft Jim Gray Systems Lab)
"Data Warehousing in the Cloud – The Death of Shared Nothing"

Post has attachment
insideHPC interviews Intel S&TC PI Jeremy Kepner of MIT Lincoln Laboratory, who founded and leads the Lab's supercomputing program, about applying supercomputing to real-world problems. Read how our research benefits from the Lab's supercomputing resources and knowledge in this recent blog post http://istc-bigdata.org/index.php/supercharging-big-data-research-with-new-englands-fastest-supercomputer/

Post has shared content
Some of our favorites are on here. See also the Intel S&TC for Big Data blog, with regular contributions from top database / Big Data researchers from +CSAIL MIT, +Brown University, +Carnegie Mellon University, +Intel, +UW (University of Washington), +Portland State University +Northwestern University University of Chicago and other academic research labs. http://istc-bigdata.org/index.php/category/istc-blog/

Post has attachment
Call for Papers for the 2017 +IEEE HPEC Conference is out, with a submission deadline of May 19. HPEC is the top conference on the convergence of High Performance and Embedded Computing. The conference is set for September 12-14, 2017, outside of Boston, Mass. See: http://www.ieee-hpec.org/index.htm
Photo

Post has attachment
New on the blog: Non-volatile memory (NVM) has been a game-changing memory technology. But it's still non-trivial for database management systems (DBMSs) to fully utilize NVM's benefits. To show NVM's possibilities for DBMSs, Intel S&TC researchers Jianhong Li Jianhong Li and Andrew Pavlo from Carnegie Mellon University turned to the open-source RocksDB key-value database system, which they enhanced to be NVM-aware. Read their post (with co-author Siying Dong of Facebook, key developer of RocksDB.)
Wait while more posts are being loaded