April 25-29, 2022- Virtual Conference
Theme: When Data Floods the Machine Room - Navigating a New Tide of Technology
Big data is certainly not new to high performance computing. In many ways, large data sets have been synonymous with HPC from the beginning -- for example, traditional modeling and simulation workloads, as well as the associated checkpointing requirements for long-running jobs, have driven various aspects of data collection, movement, compute strategy, system sizing, and storage innovation for decades. Historically, the data flood has been about volume and velocity but more recently it has become a conversation about that and much more. The more data we collect and/or generate, the more computational capacity we require to analyze it, which quickly becomes a conversation about processes and workflows: processes to massage data (analysis) to discover useful information to suggest conclusions or support decision making or to discover and communicate meaningful patterns within the data (analytics). In addition, entire industries are capitalizing on artificial intelligence to make products/manufacturing and services more effective, reliable, safer, and to enhance longevity.
In the realm of scientific computing, the theme for this year’s Salishan Conference centers on how best to prepare for and capitalize on the convergence of artificial intelligence, data analytics, and high performance computing. To address this question requires consideration at multiple levels, including the data center, hardware, software, tools, system design, flexible workload management strategies, as well as the challenges and opportunities associated with hybrid workflows. There are arguments to suggest that the ability to run multiple workloads on a single infrastructure lowers the cost for hardware, software, and maintenance while providing additional savings from reduced power and cooling costs. What are the requirements and implications for doing so? What does Data-centric design look like within the context of HPC?
Over the past decades, the annual Salishan High Speed Computing Conference has convened thought leaders of the HPC community as well as influential policymakers and program managers. Unlike an academic conference that brings together those directly engaged in research, Salishan attendees represent the broader community in government labs and agencies, industry leaders, as well as academic luminaries. Including this broader community nurtures a unique experience in which all facets of the HPC ecosystem are considered in the context of maximizing the nation’s return on investment in high performance computing. In particular, balancing the performance and cost of scientific simulations on current and future platforms has been a primary objective.
The invitation-only structure of Salishan ensured that the diversity of viewpoints and perspectives, as well as cutting-edge information within the community, could be represented, considered, and developed. As a result of its location, unique structure, the opportunity for access to peer thought leaders, and speakers and attendees chosen for their ability to influence and impact the HPC community, Salishan became and has maintained its position as one of the most sought-after invitations by HPC community leaders. In 2022, the tradition continues with the offering of 5 conference sessions. Sessions 1 through 4 consider a broad range of topics centered on the flood of data and its impacts, as well as how to navigate and capitalize on a new tide of associated technologies, while Session 5 offers a future-looking focus on advancements in Microelectronics. We are looking forward to an excellent conference.
In the realm of scientific computing, the theme for this year’s Salishan Conference centers on how best to prepare for and capitalize on the convergence of artificial intelligence, data analytics, and high performance computing. To address this question requires consideration at multiple levels, including the data center, hardware, software, tools, system design, flexible workload management strategies, as well as the challenges and opportunities associated with hybrid workflows. There are arguments to suggest that the ability to run multiple workloads on a single infrastructure lowers the cost for hardware, software, and maintenance while providing additional savings from reduced power and cooling costs. What are the requirements and implications for doing so? What does Data-centric design look like within the context of HPC?
Over the past decades, the annual Salishan High Speed Computing Conference has convened thought leaders of the HPC community as well as influential policymakers and program managers. Unlike an academic conference that brings together those directly engaged in research, Salishan attendees represent the broader community in government labs and agencies, industry leaders, as well as academic luminaries. Including this broader community nurtures a unique experience in which all facets of the HPC ecosystem are considered in the context of maximizing the nation’s return on investment in high performance computing. In particular, balancing the performance and cost of scientific simulations on current and future platforms has been a primary objective.
The invitation-only structure of Salishan ensured that the diversity of viewpoints and perspectives, as well as cutting-edge information within the community, could be represented, considered, and developed. As a result of its location, unique structure, the opportunity for access to peer thought leaders, and speakers and attendees chosen for their ability to influence and impact the HPC community, Salishan became and has maintained its position as one of the most sought-after invitations by HPC community leaders. In 2022, the tradition continues with the offering of 5 conference sessions. Sessions 1 through 4 consider a broad range of topics centered on the flood of data and its impacts, as well as how to navigate and capitalize on a new tide of associated technologies, while Session 5 offers a future-looking focus on advancements in Microelectronics. We are looking forward to an excellent conference.