Publications

Refine Results

(Filters Applied) Clear All

Performance measurements of supercomputing and cloud storage solutions

September 12, 2017
  |
Conference Paper
Author:
Michael S. Jones
…
Published in:
HPEC 2017: IEEE Conf. on High Performance Extreme Computing, 12-14 September 2017.
Topic:
high performance computing
R&D area:
R&D group:

Summary

Increasing amounts of data from varied sources, particularly in the fields of machine learning and graph analytics, are causing storage requirements to grow rapidly. A variety of technologies exist for storing and sharing these data, ranging from parallel file systems used by supercomputers to distributed block storage systems found in clouds. Relatively few comparative measurements exist to inform decisions about which storage systems are best suited for particular tasks. This work provides these measurements for two of the most popular storage technologies: Lustre and Amazon S3. Lustre is an opensource, high performance, parallel file system used by many of the largest supercomputers in the world. Amazon's Simple Storage Service, or S3, is part of the Amazon Web Services offering, and offers a scalable, distributed option to store and retrieve data from anywhere on the Internet. Parallel processing is essential for achieving high performance on modern storage systems. The performance tests used span the gamut of parallel I/O scenarios, ranging from single-client, single-node Amazon S3 and Lustre performance to a large-scale, multi-client test designed to demonstrate the capabilities of a modern storage appliance under heavy load. These results show that, when parallel I/O is used correctly (i.e., many simultaneous read or write processes), full network bandwidth performance is achievable and ranged from 10 gigabits/s over a 10 GigE S3 connection to 0.35 terabits/s using Lustre on a 1200 port 10 GigE switch. These results demonstrate that S3 is well-suited to sharing vast quantities of data over the Internet, while Lustre is well-suited to processing large quantities of data locally.
READ LESS

Summary

Increasing amounts of data from varied sources, particularly in the fields of machine learning and graph analytics, are causing storage requirements to grow rapidly. A variety of technologies exist for storing and sharing these data, ranging from parallel file systems used by supercomputers to distributed block storage systems found in...

READ MORE
Performance measurements of supercomputing and cloud storage solutions

SoK: cryptographically protected database search

May 22, 2017
  |
Conference Paper
Author:
Benjamin W. Fuller
…
Published in:
IEEE Symp. on Security and Privacy, 22-26 May 2017.
Topic:
cryptography
R&D area:
R&D group:

Summary

Protected database search systems cryptographically isolate the roles of reading from, writing to, and administering the database. This separation limits unnecessary administrator access and protects data in the case of system breaches. Since protected search was introduced in 2000, the area has grown rapidly, systems are offered by academia, start-ups, and established companies. However, there is no best protected search system or set of techniques. Design of such systems is a balancing act between security, functionality, performance, and usability. This challenge is made more difficult by ongoing database specialization, as some users will want the functionality of SQL, NoSQL, or NewSQL databases. This database evolution will continue, and the protected search community should be able to quickly provide functionality consistent with newly invented databases. At the same time, the community must accurately and clearly characterize the tradeoffs between different approaches. To address these challenges, we provide the following contributions:(1) An identification of the important primitive operations across database paradigms. We find there are a small number of base operations that can be used and combined to support a large number of database paradigms.(2) An evaluation of the current state of protected search systems in implementing these base operations. This evaluation describes the main approaches and tradeoffs for each base operation. Furthermore, it puts protected search in the context of unprotected search, identifying key gaps in functionality.(3) An analysis of attacks against protected search for different base queries.(4) A roadmap and tools for transforming a protected search system into a protected database, including an open-source performance evaluation platform and initial user opinions of protected search.
READ LESS

Summary

Protected database search systems cryptographically isolate the roles of reading from, writing to, and administering the database. This separation limits unnecessary administrator access and protects data in the case of system breaches. Since protected search was introduced in 2000, the area has grown rapidly, systems are offered by academia, start-ups...

READ MORE
SoK: cryptographically protected database search

Detecting virus exposure during the pre-symptomatic incubation period using physiological data

March 3, 2017
  |
Journal Article
Author:
Lauren Milechin
…
Published in:
Science Translational Medicine (submitted), 3 March 2017.
Topic:
detection systems
R&D area:
R&D group:

Summary

Early pathogen exposure detection allows better patient care and faster implementation of public health measures (patient isolation, contact tracing). Existing exposure detection most frequently relies on overt clinical symptoms, namely fever, during the infectious prodromal period. We have developed a robust machine learning method to better detect asymptomatic states during the incubation period using subtle, sub-clinical physiological markers. Using high-resolution physiological data from non-human primate studies of Ebola and Marburg viruses, we pre-processed the data to reduce short-term variability and normalize diurnal variations, then provided these to a supervised random forest classification algorithm. In most subjects detection is achieved well before the onset of fever; subject cross-validation lead to 52±14h mean early detection (at >0.90 area under the receiver-operating characteristic curve). Cross-cohort tests across pathogens and exposure routes also lead to successful early detection (28±16h and 43±22h, respectively). We discuss which physiological indicators are most informative for early detection and options for extending this capability to lower data resolution and wearable, non-invasive sensors.
READ LESS

Summary

Early pathogen exposure detection allows better patient care and faster implementation of public health measures (patient isolation, contact tracing). Existing exposure detection most frequently relies on overt clinical symptoms, namely fever, during the infectious prodromal period. We have developed a robust machine learning method to better detect asymptomatic states during...

READ MORE
Detecting virus exposure during the pre-symptomatic incubation period using physiological data

SIAM data mining "brings it" to annual meeting

February 17, 2017
  |
Project Report
Author:
Jeremy Kepner
…
Published in:
MIT Lincoln Laboratory Report 113081
Topic:
big data
R&D area:
R&D group:

Summary

The Data Mining Activity Group is one of SIAM's most vibrant and dynamic activity groups. To better share our enthusiasm for data mining with the broader SIAM community, our activity group organized six minisymposia at the 2016 Annual Meeting. These minisymposia included 48 talks organized by 11 SIAM members.
READ LESS

Summary

The Data Mining Activity Group is one of SIAM's most vibrant and dynamic activity groups. To better share our enthusiasm for data mining with the broader SIAM community, our activity group organized six minisymposia at the 2016 Annual Meeting. These minisymposia included 48 talks organized by 11 SIAM members.

READ MORE
SIAM data mining "brings it" to annual meeting

Learning by doing, High Performance Computing education in the MOOC era

January 17, 2017
  |
Journal Article
Author:
Julia Mullen
…
Published in:
J. Parallel Distrib. Comput., Vol. 105, July 2017, pp. 105-15.
Topic:
high performance computing
R&D area:
R&D group:

Summary

The High Performance Computing (HPC) community has spent decades developing tools that teach practitioners to harness the power of parallel and distributed computing. To create scalable and flexible educational experiences for practitioners in all phases of a career, we turn to Massively Open Online Courses (MOOCs). We detail the design of a unique self-paced online course that incorporates a focus on parallel solutions, personalization, and hands-on practice to familiarize student-users with their target system. Course material is presented through the lens of common HPC use cases and the strategies for parallelizing them. Using personalized paths, we teach researchers how to recognize the alignment between scientific applications and traditional HPC use cases, so they can focus on learning the parallelization strategies key to their workplace success. At the conclusion of their learning path, students should be capable of achieving performance gains on their HPC system.
READ LESS

Summary

The High Performance Computing (HPC) community has spent decades developing tools that teach practitioners to harness the power of parallel and distributed computing. To create scalable and flexible educational experiences for practitioners in all phases of a career, we turn to Massively Open Online Courses (MOOCs). We detail the design...

READ MORE
Learning by doing, High Performance Computing education in the MOOC era

Side channel authenticity discriminant analysis for device class identification

September 25, 2016
  |
Conference Paper
Author:
Eric A. Koziel
…
Published in:
ICIP: 2016 IEEE Int. Conf. on Image Processing, 25-28 September 2016.
Topic:
systems architecture
R&D area:
R&D group:

Summary

Counterfeit microelectronics present a significant challenge to commercial and defense supply chains. Many modern anti-counterfeit strategies rely on manufacturer cooperation to include additional identification components. We instead propose Side Channel Authenticity Discriminant Analysis (SICADA) to leverage physical phenomena manifesting from device operation to match suspect parts to a class of authentic parts. This paper examines the extent that power dissipation information can be used to separate unique classes of devices. A methodology for distinguishing device types is presented and tested on both simulation data of a custom circuit and empirical measurements of Microchip dsPIC33F microcontrollers. Experimental results show that power side channels contain significant distinguishing information to identify parts as authentic or suspect counterfeit.
READ LESS

Summary

Counterfeit microelectronics present a significant challenge to commercial and defense supply chains. Many modern anti-counterfeit strategies rely on manufacturer cooperation to include additional identification components. We instead propose Side Channel Authenticity Discriminant Analysis (SICADA) to leverage physical phenomena manifesting from device operation to match suspect parts to a class of...

READ MORE
Side channel authenticity discriminant analysis for device class identification

Designing a new high performance computing education strategy for professional scientists and engineers

September 13, 2016
  |
Conference Paper
Author:
Julia Mullen
…
Published in:
HPEC 2016: IEEE Conf. on High Performance Extreme Computing, 13-15 September 2016.
Topic:
high performance computing
R&D area:
R&D group:

Summary

For decades the High Performance Computing (HPC) community has used web content, workshops and embedded HPC scientists to enable practitioners to harness the power of parallel and distributed computing. The most successful approaches, face-to-face tutorials and embedded professionals, don't scale. To create scalable, flexible, educational experiences for practitioners in all phases of a career, from student to professional, we turn to Massively Open Online Courses (MOOCs). We detail the conversion of personalized tutorials to a selfpaced online course. In this demonstration, we highlight a course that mimics in-person tutorials by providing personalized paths through content that interleaves theory and practice, to help researchers learn key parallel computing concepts while developing familiarity with their HPC target system.
READ LESS

Summary

For decades the High Performance Computing (HPC) community has used web content, workshops and embedded HPC scientists to enable practitioners to harness the power of parallel and distributed computing. The most successful approaches, face-to-face tutorials and embedded professionals, don't scale. To create scalable, flexible, educational experiences for practitioners in all...

READ MORE
Designing a new high performance computing education strategy for professional scientists and engineers

High-throughput ingest of data provenance records in Accumulo

September 13, 2016
  |
Conference Paper
Author:
Thomas M. Moyer
…
Published in:
HPEC 2016: IEEE Conf. on High Performance Extreme Computing, 13-15 September 2016.
Topic:
supercomputing
R&D area:
R&D group:

Summary

Whole-system data provenance provides deep insight into the processing of data on a system, including detecting data integrity attacks. The downside to systems that collect whole-system data provenance is the sheer volume of data that is generated under many heavy workloads. In order to make provenance metadata useful, it must be stored somewhere where it can be queried. This problem becomes even more challenging when considering a network of provenance-aware machines all collecting this metadata. In this paper, we investigate the use of D4M and Accumulo to support high-throughput data ingest of whole-system provenance data. We find that we are able to ingest 3,970 graph components per second. Centrally storing the provenance metadata allows us to build systems that can detect and respond to data integrity attacks that are captured by the provenance system.
READ LESS

Summary

Whole-system data provenance provides deep insight into the processing of data on a system, including detecting data integrity attacks. The downside to systems that collect whole-system data provenance is the sheer volume of data that is generated under many heavy workloads. In order to make provenance metadata useful, it must...

READ MORE
High-throughput ingest of data provenance records in Accumulo

Benchmarking SciDB data import on HPC systems

September 13, 2016
  |
Conference Paper
Author:
Siddharth S. Samsi
…
Published in:
HPEC 2016: IEEE Conf. on High Performance Extreme Computing, 13-15 September 2016.
Topic:
high performance computing
R&D area:
R&D group:

Summary

SciDB is a scalable, computational database management system that uses an array model for data storage. The array data model of SciDB makes it ideally suited for storing and managing large amounts of imaging data. SciDB is designed to support advanced analytics in database, thus reducing the need for extracting data for analysis. It is designed to be massively parallel and can run on commodity hardware in a high performance computing (HPC) environment. In this paper, we present the performance of SciDB using simulated image data. The Dynamic Distributed Dimensional Data Model (D4M) software is used to implement the benchmark on a cluster running the MIT SuperCloud software stack. A peak performance of 2.2M database inserts per second was achieved on a single node of this system. We also show that SciDB and the D4M toolbox provide more efficient ways to access random sub-volumes of massive datasets compared to the traditional approaches of reading volumetric data from individual files. This work describes the D4M and SciDB tools we developed and presents the initial performance results. This performance was achieved by using parallel inserts, a in-database merging of arrays as well as supercomputing techniques, such as distributed arrays and single-program-multiple-data programming.
READ LESS

Summary

SciDB is a scalable, computational database management system that uses an array model for data storage. The array data model of SciDB makes it ideally suited for storing and managing large amounts of imaging data. SciDB is designed to support advanced analytics in database, thus reducing the need for extracting...

READ MORE
Benchmarking SciDB data import on HPC systems

In-storage embedded accelerator for sparse pattern processing

September 13, 2016
  |
Conference Paper
Author:
Sang-Woo Jun
…
Published in:
HPEC 2016: IEEE Conf. on High Performance Extreme Computing, 13-15 September 2016.
Topic:
high performance computing
R&D area:
R&D group:

Summary

We present a novel architecture for sparse pattern processing, using flash storage with embedded accelerators. Sparse pattern processing on large data sets is the essence of applications such as document search, natural language processing, bioinformatics, subgraph matching, machine learning, and graph processing. One slice of our prototype accelerator is capable of handling up to 1TB of data, and experiments show that it can outperform C/C++ software solutions on a 16-core system at a fraction of the power and cost; an optimized version of the accelerator can match the performance of a 48-core server.
READ LESS

Summary

We present a novel architecture for sparse pattern processing, using flash storage with embedded accelerators. Sparse pattern processing on large data sets is the essence of applications such as document search, natural language processing, bioinformatics, subgraph matching, machine learning, and graph processing. One slice of our prototype accelerator is capable...

READ MORE
In-storage embedded accelerator for sparse pattern processing
41-50 of 83