International Workshop on Changing Landscapes in HPC Security (CLHS), 2013Co-located with ACM HPDC, New York City, NY, USA |
---|
| The intention of this workshop is to provide an arena for researchers and practitioners to share their recent advances in HPC security, to learn from one another, and to provide a vehicle for grassroots communication between these two communities regarding their respective data and theory needs. |
---|
Overview | Providing effective and non-intrusive security within a HPC environment provides a number of challenges for both researchers and operational personnel. As the definition of what constitutes HPC expands to include cloud computing, 100G networking, cross-site integration, and web 2.0 based interfaces for job submission and reporting, the complexity of the aggregate system increases dramatically while the need for timely and accurate decision making about user activity remains unchanged. This growing complexity is balanced against a backdrop of routine user and application attacks, which remain surprisingly effective over time. This workshop will focus on the problems inherent in securing contemporary large-scale compute and storage systems. To provide some clarification we have broken this out into four general areas or questions. First is Attribution: who is doing what in terms of both process activity as well as network traffic? Second is looking beyond the interactive nodes: what is going on in the computing pool? Third involves job scheduler activity and usage: what is being run, how has it is been submitted and is this activity abnormal? Finally a more philosophical topic of why securing complex systems is so difficult and what can be done about it. While these specific areas are interesting starting points for papers and presentations, any original and interesting topic will be considered. |
---|
Format | In order to best facilitate this, a two-track approach of research papers and state of the practice write-ups will be used. We will ask participants in the research paper track to add a detailed section describing their ongoing data needs, while for the state of the practice track, we will ask that a section be added which describes possible data sharing opportunities their facility may have with individual researcher’s sites perspective. Finally the ending track will be a combined lightning round and data exchange where workshop participants can share short presentations on interesting but incomplete works as well as describe what data they have to share or need. |
---|
Topics and Details
|
Dates
|
---|
CLHS - Call for Contributions
Tentative Dates:
| Pre-Announcement |
---|
Structure of the Workshop
In order to best facilitate this workshop, a two track approach of research papers and state of the practice writeups will be used. We ask participants in the research paper track to add a detailed section describing their ongoing data needs, while for the state of the practice track, we ask that a section be added which describes possible data sharing opportunities their facility may have with individual researchers.sites perspective. Finally the ending track will be a combined lightning round and data exchange where workshop participants can share short presentations on interesting but incomplete works as well as describe what data they have to share or need.
We expect the structure of the half day workshop to resemble the following outline, with some scheduling room around the paper and SOP timing depending on the final time allocation and number of presentations.
- Opening Remarks (15 min)
- Papers presentation ( 3 x 30 min)
- break (30 min)
- SOP presentations ( 3 x 30 )
- Lightning Round/closing remarks (30 min)
Workshop Organizers
Scott Campbell
National Energy Research Scientific Computing Center (NERSC)
Lawrence Berkeley National Lab
1 Cyclotron Road Mail Stop 943-0231
510-486-6986
[email protected]
Aashish Sharma
Lawrence Berkeley National Lab
1 Cyclotron Road Mail Stop 050B-2245
510-495-2680
[email protected]
About the workshop
Providing effective and non-intrusive security within a HPC environment provides a number of challenges for both researchers and operational personnel. As the definition of what constitutes HPC expands to include cloud computing, 100G networking, cross-site integration, and web 2.0 based interfaces for job submission and reporting - the complexity of the aggregate system increases dramatically while the need for timely and accurate decision making about user activity remains unchanged. This growing complexity is balanced against a backdrop of routine user and application attacks, which remain surprisingly effective over time.
This workshop will focus on the problems inherent in securing contemporary large scale compute and storage systems. To provide some clarification we have broken this out into four general areas or questions.
- Attribution: who is doing what in terms of both process activity as well as network traffic?
- Looking beyond the interactive nodes: what is going on in the computing pool?
- Job scheduler activity and usage: what is being run, how has it is been submitted and is this activity abnormal?
- A more philosophical topic of why securing complex systems is so difficult and what can be done about it.
While these specific areas are interesting starting points for papers and presentations, any original and interesting topic will be considered. While the direct intention of this workshop is to provide an arena for researchers and practitioners to share their recent advances and to learn from one another, we believe that a grassroots effort of communication between these two communities regarding their respective data and theory needs would be tremendously fruitful.
Within the HPDC venue, there exists excellent representation for systems, application and networking research and operational deployments. With this workshop we are hoping to provide an opportunity for HPC Security research and operations to exchange meaningful information within the context of the larger conference and community. It is also worth noting that the number of security related workshops relating to HPC is exceedingly small which would prove beneficial in attendance and submission.
How and What to Submit
Stay tunned!!!