High Performance Computing (HPC) in the Cloud - By Amazon AWS

High Performance Computing (HPC) in the Cloud
Guy Ernest, Solution Architect. Amazon AWS

Thousands of companies today are using data to differentiate their products and disrupt a growing number of industries, including media/advertising, retail, gaming, health care, and financial services. High Performance Computing (HPC) allows scientists and engineers to solve complex science, engineering and business problems using applications that require high bandwidth, low latency networking, and very high compute capabilities.

Learn how the AWS cloud can cost effectively provide the scalable computing resources, storage services, and analytics tools required for anyone to participate in these technologies.

Key topics at this meet-up will include:
· High performance parallel computing with AWS Cluster Compute Instance family.
· Hadoop and scale out computation with Amazon Elastic MapReduce (EMR)
· Customer case studies

Amazon HPC Cloud Compute Resource Usage at the ACLP for the NIST OpenKWS-2013 Evaluation
During April-May of 2013 the ACLP participated in the NIST OpenKWS13 evaluation. This activity required the rapid construction of a keyword spotting system, as well as running an extensive experiment on a large amount of speech data in a very limited time frame. The situation precluded the usage of locally acquired computation resources, as a large amount of computing power was needed immediately for a limited amount of time. The obvious solution was to use public cloud compute resources, and the choice was made to use the AWS ECC solution. As the evaluation was CPU intensive, it was decided to use the high-end cluster-compute nodes offered by Amazon. In all, 8 instances were utilized, each with dual octa-core CPU, and ~60 GB of RAM. This was performed over the course of several weeks, for a total of ~750 hours of run-time

Presenter: Prof. Ami Moyal
Head of Electrical Engineering Department at Afeka Tel Aviv Academic College of Engineering

Head of ACLP - Afeka Center for Language Processing at Afeka Tel Aviv Academic College of Engineering

Who Should Attend?
Engineers, architects, product managers, data scientists, high performance computing specialists, grid users and researchers from industry and academia, along with technically-minded business stakeholders looking to put data to work for their organization.