Our presentations will discuss data management concepts critical to facilitating the use of high-performance computing. We hope to see you at one of our scheduled presentations or in the Exhibit area. Register here to attend.
Presented by Bob Murphy
As HPC workloads continue to grow in size and complexity, efficient data management becomes
increasingly crucial for the overall performance of the system. This presentation explores
the impact of data management on HPC workloads and how critical it is becoming for efficient
HPC.
Groundbreaking discoveries often come with their own unique data management challenges. This
includes the need to handle large volumes of data, manage data access and sharing, and
ensure data consistency and integrity. We will discuss different approaches to data
management in HPC systems, including distributed file systems, parallel I/O, and data
caching. The datasets associated with these research initiatives are ever-growing, yet
nonetheless require effective data and storage management. Data Managers are faced with
tradeoffs that balance the interests of researchers with that of infrastructure managers
mindful of long-term data access, security, and storage costs. The importance of achieving
the right data management strategy becomes paramount as the size of HPC datasets continues
to increase.
Join us as we discuss the impact of data management on HPC workflows, survey some of the
current research trends and future directions, and explore real-world use cases and best
practices from organizations optimizing their data management in support of breakthrough
research. The session will explore data management for immediate computational needs as well
as alternatives for long-term data access, management, and preservation. This is an
interactive session where we invite the audience to share best practices.
Marcel Fallet, STEM Technical Leader at the National Security Agency, Washington, DC.
Volker Eyrich, Google Customer Engineer, scientist and research computing specialist, Austin, TX.
Sara Willis, Computing Sciences Researcher at the University of Arizona, Tucson, AZ.
Graham Beasley, Chief Operating Officer of the international software company Arcitecta, Louisville, CO.
A student career panel where you will hear from HPC professionals about how they entered the HPC workforce and what they are looking for when hiring new employees. This will be a one hour session broken into presentations, with time allocated for questions.
Presented by Graham Beasley
High-performance computing (HPC) systems are often used to process vast amounts of data in
scientific research, engineering, and other fields. As data sets become increasingly complex
and large, managing data becomes a significant challenge for HPC users. Metadata is an
essential technology that can help alleviate this burden by providing a way to describe,
organize and manage data.
The importance of metadata in HPC moving forward cannot be overstated. It is a fundamental
technology that helps researchers better understand their data, making it easier to search,
access, and analyze. Metadata can describe the structure, content, and context of data,
enabling researchers to quickly find and access the information they need without having to
search through vast amounts of data. It can also be used to ensure the quality and accuracy
of data and track the data’s provenance and usage.
Metadata is especially useful in large-scale scientific research projects where data is
shared among multiple researchers and institutions. Metadata can help researchers understand
how the data was generated, what assumptions were made, and what processing steps were
taken. It can also be used to ensure data consistency and accuracy across different
locations, allowing researchers to collaborate more effectively.
And metadata saves money. HPC systems often require massive amounts of storage, which can be
difficult to manage without an efficient and scalable data management solution. Metadata can
be used to organize data and make it more manageable, reducing the amount of storage
required. This can lead to significant cost savings for institutions, as well as improved
data management and processing.
