11/08/2016

Reservoir Labs at Supercomputing 2016 (SC16)

Reservoir Labs will be participating in Supercomputing 2016 (SC16) on Friday, November 11 through Thursday, November 17, in Salt Lake City, UT. Please join us at the following venues or contact us for a private briefing.

SCinet The Fastest Network Connecting the Fastest Computers
Annually, Supercomputing hosts one of the most powerful and advanced networks in the world —SCinet. SCinet brings to life a high-capacity network that supports the revolutionary applications and experiments that are a hallmark of the SC conference. SCinet will link the convention center to research and commercial networks around the world. In doing so, SCinet serves as the platform for exhibitors to demonstrate the advanced computing resources of their home institutions and elsewhere by supporting a wide variety of bandwidth-driven applications including supercomputing and cloud computing. Reservoir Labs supports the Network Security capabilities of SCinet with R-Scope® Advanced Threat Solution and ENSIGN Hypergraph analytics.

SCinet Demonstrations are available through NRE, Network Research Exhibition, throughout the week.
Each year the SCinet NRE showcases the leading network-based solutions that were created to deliver the highest performance available within the greatest security footprint. The goal of the NRE is to showcase technologies that will impact HPC in general and SCinet in particular. Topics for SC16’s NRE demonstrations and experiments range from software-defined networking (SDN) to security/encryption and resilience.

Reservoir Labs will demonstrate the fusion of two of its leading technologies – 1) R-Scope, a high-performance cyber security appliance enabling deep network visibility, advanced situational awareness, and real-time security event detection by extracting cyber-relevant data from network traffic, and 2) ENSIGN, a high-performance Big Data analysis tool that provides the fastest and most scalable tensor analysis routines to reveal interesting patterns and discover subtle undercurrents and deep cross-dimensional correlations in data. Our objective is to demonstrate the effectiveness of ENSIGN analytics in an operational cyber security setup to extract anomalous patterns of network traffic, detect alarming behaviors, and provide actionable insights into network data, by analyzing network metadata output logs from the R-Scope systems.

Please contact Reservoir Labs if you would like to schedule a demonstration.

HP-CAST27  HP Consortium for Advanced Scientific and Technical Computing.
Friday – Saturday, November 11-12, at the Salt Lake Marriott Downtown at City Creek

High Performance Cybersecurity Analytics: SCinet, a Case About Detecting Vulnerability at Scale
Ms. Alison Ryan and Dr. Richard Lethin will share business and technical perspectives on HPC Network Security and Analytics and solutions deployed on HPE compute infrastructure.

The most damaging cyber security attacks to both commercial and government entities are those patterns we’ve not yet seen and that target our most valuable assets.  Hunting for this type of assault requires highly scalable and reliable sensors and an analytic platform that processes multi-dimensional, structured and unstructured data, and makes it easier for security professionals to hunt at scale.  We will share how integrating HPE Systems with novel network data extraction and machine learning tools is being used by Fortune 500 firms and government agencies nationwide.  In the context of SC16, we are extending this solution to hunt at extreme scale.  This comprehensive solution is part of the security infrastructure for SCinet, the large-scale research network stood up each year in support of SC.  We will demonstrate  how to combine deep network visibility with machine learning, to identify otherwise unseen attacks amongst  billions of flows, which  reveal actionable patterns that can be easily used by security practitioners.  

Please note HP-CAST27 Agenda.

INDIS Innovating the Network for Data-Intensive Science
Sunday Nov. 13, 9:20 AM at Salt Palace Convention Center, Room 251-A

Reservoir Labs: High-Performance Algorithms and Data Structures to Catch Elephant Flows
Dr. Jordi Ros-Giralt

As part of the SCinet Network Research Exposition (NRE), Reservoir Labs will demonstrate rHNTES/FlowTec, a new mathematical framework, algorithm and a high-performance implementation to detect very large flows – known as elephant flows — at very high-speed rates.  rHNTES/FlowTec helps ensure the quality of service (QoS) required by the users of high speed networks. Our technology effectively detects the presence of elephant flows on high-speed networks such as the ESnet —  the large flows carrying big data from scientific experiments — and enables QoS mechanisms for routing to protect the numerous smaller delay-sensitive flows.  rHNTES/FlowTec has been developed by Reservoir Labs in partnership with the University of Virginia.

This third annual INDIS workshop is part of the SC16 Tech Program. INDIS showcases both Demonstrations and Technical Papers highlighting important developments in high-performance networking. Reservoir Labs invites you to attend our NRE Theme Presentation and the SCinet innovation panel that will take place during the SC INDIS Workshop. Dr. Jordi Ros-Giralt will present our Network Research Exhibition (NRE) Demo including rHNTES/FlowTec and our cyber security for SCinet.

ESPT Extreme-Scale Programming Tools Workshop
Sunday November 13, 16:45-17:10 Salt Lake City Conversion Center, Room 155-E

Reservoir Labs: Automatic Code Generation for an Asynchronous Task-Based Runtime
Dr. Muthu Baskaran

Dr. Baskaran will present Reservoir’s technology for automatic parallelization and optimization of scientific computing kernels to event-driven-task (EDT) runtimes.  Reservoir’s new mixed static-dynamic compiler parallelization algorithms can now generate (1) dynamically self-unfolding event-driven task programs, with (2) the associated optimized explicit asynchronous communications and (3) the explicit automatic memory management of data blocks.  EDT execution substantially increases concurrency by removing redundant synchronization and also by maximizing the opportunity for dynamic scheduling within the remaining needed constraints.  This enables both performance and power savings on current high performance computing systems, and with even greater benefits on anticipated exascale systems. Reservoir’s EDT program generation capability expands the applicability of the EDT technique, removing the need for complex manual programming to get the benefit of EDT and also by enabling the generation of complex EDT mappings for extreme performance, which would be beyond the capability of manual programming. Reservoir’s dynamic technique eliminates the need seen in other systems, to fully form the task graph at startup.  This approach reduces startup time and memory requirements. We will demonstrate the effectiveness of our technique and the performance improvements obtained using benchmarks and exascale proxy applications, targeted toward fast implementations of the Open Community Runtime (OCR).

SCinet Briefing—Security in a High Performance Environment
(aka, Lessons from the Trenches)
Tuesday, November 15 at 1:30 pm, at the Salt Palace Convention Center, SC16 – room 250-D

During this session, thought leaders from the SCinet team (including Reservoir engineers) will highlight the Network Security infrastructure created for SCinet and share insights gathered from architecting, supporting and securing SC16’s communication backbone. Following presentations, time has been allocated for interactive discussion.

IA3 Sixth Workshop on Irregular Applications: Architectures and Algorithms.
Sunday, November 13, 2016, Salt Lake City Convention Center, Room 251- Emu presentation at 2:40

Reservoir’s client, EMU Technology, will present work on an exciting new supercomputer architecture specifically designed for irregular applications such as Graphs and Machine Learning.  Reservoir authors contributed specifically to the LLVM-based Cilk compiler, the overall programming model, and co-design of the architecture.  The paper, titled “Highly Scalable Near Memory Processing with Migrating Threads on the Emu System Architecture” has lead authors from Emu: Timothy Dysart, Peter Kogge, and Martin Deneroff.  Reservoir authors on the paper are Preston Briggs, Richard Lethin, and John Ruttenberg.