Publications

Refine Results

(Filters Applied) Clear All

By

75,000,000,000 streaming inserts/second using hierarchical hypersparse GraphBLAS matrices

May 18, 2020
  |
Conference Paper
Author:
Jeremy Kepner
…
Published in:
2020 IEEE Intl. Parallel and Distributed Processing Symp. Workshops, IPDPSW, 18-22 May 2020.
Topic:
supercomputing
R&D area:
R&D group:

Summary

The SuiteSparse GraphBLAS C-library implements high performance hypersparse matrices with bindings to a variety of languages (Python, Julia, and Matlab/Octave). GraphBLAS provides a lightweight in-memory database implementation of hypersparse matrices that are ideal for analyzing many types of network data, while providing rigorous mathematical guarantees, such as linearity. Streaming updates of hypersparse matrices put enormous pressure on the memory hierarchy. This work benchmarks an implementation of hierarchical hypersparse matrices that reduces memory pressure and dramatically increases the update rate into a hypersparse matrices. The parameters of hierarchical hypersparse matrices rely on controlling the number of entries in each level in the hierarchy before an update is cascaded. The parameters are easily tunable to achieve optimal performance for a variety of applications. Hierarchical hypersparse matrices achieve over 1,000,000 updates per second in a single instance. Scaling to 31,000 instances of hierarchical hypersparse matrices arrays on 1,100 server nodes on the MIT SuperCloud achieved a sustained update rate of 75,000,000,000 updates per second. This capability allows the MIT SuperCloud to analyze extremely large streaming network data sets.
READ LESS

Summary

The SuiteSparse GraphBLAS C-library implements high performance hypersparse matrices with bindings to a variety of languages (Python, Julia, and Matlab/Octave). GraphBLAS provides a lightweight in-memory database implementation of hypersparse matrices that are ideal for analyzing many types of network data, while providing rigorous mathematical guarantees, such as linearity. Streaming updates...

READ MORE
75,000,000,000 streaming inserts/second using hierarchical hypersparse GraphBLAS matrices

Large scale parallelization using file-based communications

September 24, 2019
  |
Conference Paper
Author:
Chansup Byun
…
Published in:
2019 IEEE High Performance Computing Conf., HPEC, 24-26 September 2019.
Topic:
high performance computing
R&D area:
R&D group:

Summary

In this paper, we present a novel and new file-based communication architecture using the local filesystem for large scale parallelization. This new approach eliminates the issues with filesystem overload and resource contention when using the central filesystem for large parallel jobs. The new approach incurs additional overhead due to inter-node message file transfers when both the sending and receiving processes are not on the same node. However, even with this additional overhead cost, its benefits are far greater for the overall cluster operation in addition to the performance enhancement in message communications for large scale parallel jobs. For example, when running a 2048-process parallel job, it achieved about 34 times better performance with MPI_Bcast() when using the local filesystem. Furthermore, since the security for transferring message files is handled entirely by using the secure copy protocol (scp) and the file system permissions, no additional security measures or ports are required other than those that are typically required on an HPC system.
READ LESS

Summary

In this paper, we present a novel and new file-based communication architecture using the local filesystem for large scale parallelization. This new approach eliminates the issues with filesystem overload and resource contention when using the central filesystem for large parallel jobs. The new approach incurs additional overhead due to inter-node...

READ MORE
Large scale parallelization using file-based communications

Streaming 1.9 billion hyperspace network updates per second with D4M

September 24, 2019
  |
Conference Paper
Author:
Jeremy Kepner
…
Published in:
2019 IEEE High Performance Computing Conf., HPEC, 24-26 September 2019.
Topic:
high performance computing
R&D area:
R&D group:

Summary

The Dynamic Distributed Dimensional Data Model (D4M) library implements associative arrays in a variety of languages (Python, Julia, and Matlab/Octave) and provides a lightweight in-memory database implementation of hypersparse arrays that are ideal for analyzing many types of network data. D4M relies on associative arrays which combine properties of spreadsheets, databases, matrices, graphs, and networks, while providing rigorous mathematical guarantees, such as linearity. Streaming updates of D4M associative arrays put enormous pressure on the memory hierarchy. This work describes the design and performance optimization of an implementation of hierarchical associative arrays that reduces memory pressure and dramatically increases the update rate into an associative array. The parameters of hierarchical associative arrays rely on controlling the number of entries in each level in the hierarchy before an update is cascaded. The parameters are easily tunable to achieve optimal performance for a variety of applications. Hierarchical arrays achieve over 40,000 updates per second in a single instance. Scaling to 34,000 instances of hierarchical D4M associative arrays on 1,100 server nodes on the MIT SuperCloud achieved a sustained update rate of 1,900,000,000 updates per second. This capability allows the MIT SuperCloud to analyze extremely large streaming network data sets.
READ LESS

Summary

The Dynamic Distributed Dimensional Data Model (D4M) library implements associative arrays in a variety of languages (Python, Julia, and Matlab/Octave) and provides a lightweight in-memory database implementation of hypersparse arrays that are ideal for analyzing many types of network data. D4M relies on associative arrays which combine properties of spreadsheets...

READ MORE
Streaming 1.9 billion hyperspace network updates per second with D4M

A billion updates per second using 30,000 hierarchical in-memory D4M databases

January 24, 2019
  |
Conference Paper
Author:
Jeremy Kepner
…
Published in:
North East Database Day, NEDB Day, 24 January 2019.
Topic:
high performance computing
R&D area:
R&D group:

Summary

Analyzing large scale networks requires high performance streaming updates of graph representations of these data. Associative arrays are mathematical objects combining properties of spreadsheets, databases, matrices, and graphs, and are well-suited for representing and analyzing streaming network data. The Dynamic Distributed Dimensional Data Model (D4M) library implements associative arrays in a variety of languages (Python, Julia, and Matlab/Octave) and provides a lightweight in-memory database. Associative arrays are designed for block updates. Streaming updates to a large associative array requires a hierarchical implementation to optimize the performance of the memory hierarchy. Running 34,000 instances of a hierarchical D4M associative arrays on 1,100 server nodes on the MIT SuperCloud achieved a sustained update rate of 1,900,000,000 updates per second. This capability allows the MIT SuperCloud to analyze extremely large streaming network data sets.
READ LESS

Summary

Analyzing large scale networks requires high performance streaming updates of graph representations of these data. Associative arrays are mathematical objects combining properties of spreadsheets, databases, matrices, and graphs, and are well-suited for representing and analyzing streaming network data. The Dynamic Distributed Dimensional Data Model (D4M) library implements associative arrays in...

READ MORE
A billion updates per second using 30,000 hierarchical in-memory D4M databases

Hyperscaling internet graph analysis with D4M on the MIT SuperCloud

September 26, 2018
  |
Conference Paper
Author:
Vijay N. Gadepally
…
Published in:
IEEE High Performance Extreme Computing Conf., HPEC, 25-27 September 2018.
Topic:
high performance computing
R&D area:
R&D group:

Summary

Detecting anomalous behavior in network traffic is a major challenge due to the volume and velocity of network traffic. For example, a 10 Gigabit Ethernet connection can generate over 50 MB/s of packet headers. For global network providers, this challenge can be amplified by many orders of magnitude. Development of novel computer network traffic analytics requires: high level programming environments, massive amount of packet capture (PCAP) data, and diverse data products for "at scale" algorithm pipeline development. D4M (Dynamic Distributed Dimensional Data Model) combines the power of sparse linear algebra, associative arrays, parallel processing, and distributed databases (such as SciDB and Apache Accumulo) to provide a scalable data and computation system that addresses the big data problems associated with network analytics development. Combining D4M with the MIT SuperCloud manycore processors and parallel storage system enables network analysts to interactively process massive amounts of data in minutes. To demonstrate these capabilities, we have implemented a representative analytics pipeline in D4M and benchmarked it on 96 hours of Gigabit PCAP data with MIT SuperCloud. The entire pipeline from uncompressing the raw files to database ingest was implemented in 135 lines of D4M code and achieved speedups of over 20,000.
READ LESS

Summary

Detecting anomalous behavior in network traffic is a major challenge due to the volume and velocity of network traffic. For example, a 10 Gigabit Ethernet connection can generate over 50 MB/s of packet headers. For global network providers, this challenge can be amplified by many orders of magnitude. Development of...

READ MORE
Hyperscaling internet graph analysis with D4M on the MIT SuperCloud

Interactive supercomputing on 40,000 cores for machine learning and data analysis

September 25, 2018
  |
Conference Paper
Author:
Albert I. Reuther
…
Published in:
IEEE High Performance Extreme Computing Conf., HPEC, 25-27 September 2018.
Topic:
high performance computing
R&D area:
R&D group:

Summary

Interactive massively parallel computations are critical for machine learning and data analysis. These computations are a staple of the MIT Lincoln Laboratory Supercomputing Center (LLSC) and has required the LLSC to develop unique interactive supercomputing capabilities. Scaling interactive machine learning frameworks, such as TensorFlow, and data analysis environments, such as MATLAB/Octave, to tens of thousands of cores presents many technical challenges – in particular, rapidly dispatching many tasks through a scheduler, such as Slurm, and starting many instances of applications with thousands of dependencies. Careful tuning of launches and prepositioning of applications overcome these challenges and allow the launching of thousands of tasks in seconds on a 40,000-core supercomputer. Specifically, this work demonstrates launching 32,000 TensorFlow processes in 4 seconds and launching 262,000 Octave processes in 40 seconds. These capabilities allow researchers to rapidly explore novel machine learning architecture and data analysis algorithms.
READ LESS

Summary

Interactive massively parallel computations are critical for machine learning and data analysis. These computations are a staple of the MIT Lincoln Laboratory Supercomputing Center (LLSC) and has required the LLSC to develop unique interactive supercomputing capabilities. Scaling interactive machine learning frameworks, such as TensorFlow, and data analysis environments, such as...

READ MORE
Interactive supercomputing on 40,000 cores for machine learning and data analysis

A parallel implementation of FANO using OpenMP and MPI

September 25, 2018
  |
Conference Paper
Author:
Plamen Krastev
…
Published in:
IEEE High Performance Extreme Computing Conf., HPEC, 25-27 September 2018.
Topic:
high performance computing
R&D area:
R&D group:

Summary

We present a parallel implementation of the Fast Accurate NURBS Optimization (FANO) program using OpenMP and MPI. The software is used for designing imaging freeform optical systems comprised of NURBS surfaces. An important step in the design process is the optimization of the shape and position of the optical surfaces within the optical system. FANO uses the Levenberg-Marquardt (LM) algorithm for minimization of the merit function. The parallelization of the code is achieved without modifying readily available commercial or open source implementations of the LM algorithm. Instead, MPI instructions are being used to distribute the computation of the Jacobian over multiple nodes, each of which performs the computationally intensive task of raytracing. The results from the raytracing are collected on the master and used for calculating the values of the variable parameters for the next iteration. Speed increases of ~100x and more are possible when running on the cluster of the MIT Lincoln Laboratory Super Computing Center (LLSC).
READ LESS

Summary

We present a parallel implementation of the Fast Accurate NURBS Optimization (FANO) program using OpenMP and MPI. The software is used for designing imaging freeform optical systems comprised of NURBS surfaces. An important step in the design process is the optimization of the shape and position of the optical surfaces...

READ MORE
A parallel implementation of FANO using OpenMP and MPI

Measuring the impact of Spectre and Meltdown

September 25, 2018
  |
Conference Paper
Author:
Andrew J. Prout
…
Published in:
IEEE High Performance Extreme Computing Conf., HPEC, 25-27 September 2018.
Topic:
high performance computing
R&D area:
R&D group:

Summary

The Spectre and Meltdown flaws in modern microprocessors represent a new class of attacks that have been difficult to mitigate. The mitigations that have been proposed have known performance impacts. The reported magnitude of these impacts varies depending on the industry sector and expected workload characteristics. In this paper, we measure the performance impact on several workloads relevant to HPC systems. We show that the impact can be significant on both synthetic and realistic workloads. We also show that the performance penalties are difficult to avoid even in dedicated systems where security is a lesser concern.
READ LESS

Summary

The Spectre and Meltdown flaws in modern microprocessors represent a new class of attacks that have been difficult to mitigate. The mitigations that have been proposed have known performance impacts. The reported magnitude of these impacts varies depending on the industry sector and expected workload characteristics. In this paper, we...

READ MORE
Measuring the impact of Spectre and Meltdown

Lessons learned from a decade of providing interactive, on-demand high performance computing to scientists and engineers

June 28, 2018
  |
Conference Paper
Author:
Julia Mullen
…
Published in:
ISC High Performance Int. Workshops, Lecture Notes in Computer Science (LNSC 11203), 28 June 2018, pp. 655-68.
Topic:
high performance computing
R&D area:
R&D group:

Summary

For decades, the use of HPC systems was limited to those in the physical sciences who had mastered their domain in conjunction with a deep understanding of HPC architectures and algorithms. During these same decades, consumer computing device advances produced tablets and smartphones that allow millions of children to interactively develop and share code projects across the globe. As the HPC community faces the challenges associated with guiding researchers from disciplines using high productivity interactive tools to effective use of HPC systems, it seems appropriate to revisit the assumptions surrounding the necessary skills required for access to large computational systems. For over a decade, MIT Lincoln Laboratory has been supporting interactive, on demand high performance computing by seamlessly integrating familiar high productivity tools to provide users with an increased number of design turns, rapid prototyping capability, and faster time to insight. In this paper, we discuss the lessons learned while supporting interactive, on-demand high performance computing from the perspectives of the users and the team supporting the users and the system. Building on these lessons, we present an overview of current needs and the technical solutions we are building to lower the barrier to entry for new users from the humanities, social, and biological sciences.
READ LESS

Summary

For decades, the use of HPC systems was limited to those in the physical sciences who had mastered their domain in conjunction with a deep understanding of HPC architectures and algorithms. During these same decades, consumer computing device advances produced tablets and smartphones that allow millions of children to interactively...

READ MORE
Lessons learned from a decade of providing interactive, on-demand high performance computing to scientists and engineers

Benchmarking data analysis and machine learning applications on the Intel KNL many-core processor

September 12, 2017
  |
Conference Paper
Author:
Chansup Byun
…
Published in:
HPEC 2017: IEEE Conf. on High Performance Extreme Computing, 12-14 September 2017.
Topic:
machine learning
R&D area:
R&D group:

Summary

Knights Landing (KNL) is the code name for the second-generation Intel Xeon Phi product family. KNL has generated significant interest in the data analysis and machine learning communities because its new many-core architecture targets both of these workloads. The KNL many-core vector processor design enables it to exploit much higher levels of parallelism. At the Lincoln Laboratory Supercomputing Center (LLSC), the majority of users are running data analysis applications such as MATLAB and Octave. More recently, machine learning applications, such as the UC Berkeley Caffe deep learning framework, have become increasingly important to LLSC users. Thus, the performance of these applications on KNL systems is of high interest to LLSC users and the broader data analysis and machine learning communities. Our data analysis benchmarks of these application on the Intel KNL processor indicate that single-core double-precision generalized matrix multiply (DGEMM) performance on KNL systems has improved by ~3.5x compared to prior Intel Xeon technologies. Our data analysis applications also achieved ~60% of the theoretical peak performance. Also a performance comparison of a machine learning application, Caffe, between the two different Intel CPUs, Xeon E5 v3 and Xeon Phi 7210, demonstrated a 2.7x improvement on a KNL node.
READ LESS

Summary

Knights Landing (KNL) is the code name for the second-generation Intel Xeon Phi product family. KNL has generated significant interest in the data analysis and machine learning communities because its new many-core architecture targets both of these workloads. The KNL many-core vector processor design enables it to exploit much higher...

READ MORE
Benchmarking data analysis and machine learning applications on the Intel KNL many-core processor
1-10 of 29