Sign up for our newsletter and get the latest HPC news and analysis.
Send me information from insideHPC:


What would you do with a Million cores?

Ian Colle from AWS gave this talk at the Univa Breakfast Briefing at ISC 2018. “To demonstrate the unique ability to run very large enterprise HPC clusters and workloads, Univa leveraged AWS to deploy 1,015,022 cores in a single Univa Grid Engine cluster to showcase the advantages of running large-scale electronic design automation (EDA) workloads in the cloud. The cluster was built in approximately 2.5 hours using Navops Launch automation and comprised more than 55,000 AWS instances in 3 availability zones, 16 different instance types and leveraged AWS Spot Fleet technology to maximize the rate at which Amazon EC2 hosts were launched while enabling capacity and costs to be managed according to policy.”

Univa: Optimizing On-Premise Clusters and Migration to the Cloud

In this video from ISC 2018, Rob Lalonde describes how Univa products optimize on-premise clusters and migration to the Cloud. “Univa is the leading independent provider of software-defined computing infrastructure and workload orchestration solutions. Univa’s intelligent cluster management software increases efficiency while accelerating enterprise migration to hybrid clouds. We help hundreds of companies to manage thousands of applications and run billions of tasks every day.”

Univa Deploys Million-core Grid Engine Cluster on AWS

To demonstrate the unique ability to run very large enterprise HPC clusters and workloads, Univa leveraged AWS to deploy 1,015,022 cores in a single Univa Grid Engine cluster to showcase the advantages of running large-scale electronic design automation (EDA) workloads in the cloud. The cluster was built in approximately 2.5 hours using Navops Launch automation and comprised more than 55,000 AWS instances in 3 availability zones, 16 different instance types and leveraged AWS Spot Fleet technology to maximize the rate at which Amazon EC2 hosts were launched while enabling capacity and costs to be managed according to policy.

Univa to host Breakfast Briefing on HPC in the Cloud at ISC 2018

Univa will host a Breakfast Briefing on HPC in the Cloud at ISC 2018. “You are invited to attend this briefing on HPC in the Cloud at ISC 2018. A panel of product executives from Amazon Web Services and Nvidia will join Univa to share real-world use cases and discuss best practices for accelerating HPC cloud migration. Please join us and see how to increase the efficiency of on-premise data centers, and address peak needs with hybrid cloud bursting.”

AWS Announces EC2 Bare Metal instances and C5D instances with local SSD Storage

HPC in the Cloud got a boost this week with this announcement of two updates to the Amazon EC2 instance family. EC2 Bare Metal instances provide customers’ applications with direct access to the processor and memory resources of the underlying server. C5D instances with local SSD storage are designed for applications that benefit from high-speed, ultra-low latency local storage, such as video encoding, manipulation, and other forms of media processing that often include large numbers of reads and writes to temporary storage.

Universities step up to Cloud Bursting

In this special guest feature, Mahesh Pancholi from OCF writes that many of universities are now engaging in cloud bursting and are regularly taking advantage of public cloud infrastructures that are widely available from large companies like Amazon, Google and Microsoft. “By bursting into the public cloud, the university can offer the latest and greatest technologies as part of its Research Computing Service for all its researchers.”

Amazon and Libfabric: A case study in flexible HPC Infrastructure

Brian Barrett from Amazon gave this talk at the 2018 OpenFabrics Workshop. “As network performance becomes a larger bottleneck in application performance, AWS is investing in improving HPC network performance. Our initial investment focused on improving performance in open source MPI implementations, with positive results. Recently, however, we have pivoted to focusing on using libfabric to improve point to point performance.”

Amazon SageMaker goes for “Infinitely Scalable” Machine Learning

Over at the All Things Distributed blog, Werner Vogels writes that the new Amazon SageMaker is designed for building machine learning algorithms that can handle an infinite amount of data. “To handle unbounded amounts of data, our algorithms adopt a streaming computational model. In the streaming model, the algorithm only passes over the dataset one time and assumes a fixed-memory footprint. This memory restriction precludes basic operations like storing the data in memory, random access to individual records, shuffling the data, reading through the data several times, etc.”

Big 3 Cloud Providers join with NSF to Support Data Science

“NSF’s participation with major cloud providers is an innovative approach to combining resources to better support data science research,” said Jim Kurose, assistant director of NSF for Computer and Information Science and Engineering (CISE). “This type of collaboration enables fundamental research and spurs technology development and economic growth in areas of mutual interest to the participants, driving innovation for the long-term benefit of our nation.”

Video: How to Get the HPC Best-in-class Performance via Intel Xeon Skylake

“HPC Cloud services built on the latest Intel architecture, Skylake Xeon processor, are now powering the C5 compute intensive instance at AWS and can serve as your next-generation HPC platform. Hear how customers are starting to consider hybrid strategies to increase productivity and lower their capital expenditure and maintenance costs. Also learn how to adapt this model to meet the increasing HPC and data analytics needs for your applications with the new technologies incorporated into the platform.”