BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T143036Z
LOCATION:B313
DTSTART;TZID=America/New_York:20241117T090000
DTEND;TZID=America/New_York:20241117T173000
UID:submissions.supercomputing.org_SC24_sess732@linklings.com
SUMMARY:CANOPIE-HPC: 6th International Workshop on Containers and New Orch
 estration Paradigms for Isolated Environments in HPC
DESCRIPTION:The ongoing revolution enabled via containerization, virtualiz
 ation, and new orchestration models has dramatically changed how applicati
 ons and services are delivered and managed across the computing industry. 
 This revolution has established a new ecosystem of tools and techniques wi
 th new, flexible, and agile approaches, and continues to gain traction in 
 the HPC community. In addition to HPC-optimized container runtimes, emergi
 ng technologies like Kubernetes create a new set of opportunities and chal
 lenges. While adoption is growing, questions regarding best practices, fou
 ndational concepts, tools, and standards remain. Our goal is to promote th
 e adoption of these tools and introspect the impact of this new ecosystem 
 on HPC use cases. This workshop serves as a key venue for presenting late-
 breaking research, sharing experiences and best practices, and fostering c
 ollaboration in this field. Our sixth workshop iteration will continue to 
 emphasize real-world experiences and challenges in adopting and optimizing
  these new approaches for HPC.\n\nZero-Consistency Root Emulation for Unpr
 ivileged Container Build\n\nDo Linux distribution package managers need th
 e privileged operations they request to actually happen? Apparently not, a
 t least when building container images for HPC applications. We use this o
 bservation to implement a root emulation mode using a Linux seccomp filter
  that intercepts some privileged...\n\n\nReid Priedhorsky, Michael Jenning
 s, and Megan Phinney (Los Alamos National Laboratory (LANL))\n------------
 ---------\nComparative Analysis of a Containerized Compilation Process Ver
 sus Bare-Metal Compilation Runs\n\nThe National Oceanic and Atmospheric Ad
 ministration (NOAA) strives to understand and predict changes in the envir
 onment through improving climate modeling. These models can be tested for 
 reproducibility using tools in the Flexible Modeling System (FMS) Runtime 
 Environment (FRE) framework. The FRE ec...\n\n\nDana Singh (Science Applic
 ations International Corporation (SAIC)) and Thomas Robinson (Geophysical 
 Fluid Dynamics Laboratory (NOAA))\n---------------------\nCANOPIE-HPC — Af
 ternoon Introduction\n\nAlberto Madonna (Swiss National Supercomputing Cen
 tre (CSCS)) and Claudia Misale (IBM TJ Watson Research Center)\n----------
 -----------\nMorning Break\n---------------------\nCANOPIE-HPC — Conclusio
 ns and Farewell\n\nAlberto Madonna (Swiss National Supercomputing Centre (
 CSCS)), Claudia Misale, Richard Shane Canon (Lawrence Berkeley National La
 boratory (LBNL)), and Andrew Younge (Sandia National Laboratories)\n------
 ---------------\nSlinky: The Missing Link Between Slurm and Kubernetes\n\n
 AI and ML engineers are often unfamiliar with traditional HPC environments
 . They rely on cloud native systems like Kubernetes to abstract complex sy
 stem management. These platforms lack the fine-grained resource control an
 d advanced scheduling features crucial for HPC/AI/ML workloads. Slurm, an 
 open...\n\n\nTim Wickberg and Skyler Malinowski (SchedMD LLC)\n-----------
 ----------\nK-Foundry: Using Kubernetes-Like Control Planes with Custom Re
 source Definitions to Deploy Containerized HPC Applications Across Multipl
 e Computing Platforms\n\nWe present K-Foundry, a framework that enables th
 e integration of Simple Linux Utility for Resource Management (SLURM) with
  Kubernetes (K8s) via a Kubernetes-like Control Plane (KCP). Our implement
 ation seamlessly enables a unified communication and scheduling layer for 
 a fleet of multiple diverse c...\n\n\nAngel M. Beltre (Sandia National Lab
 oratories); Carlos Eduardo Arango Gutierrez (NVIDIA Corporation); and Sylv
 ain Bernard, Kevin T. Pedretti, John M. Linebarger, Stephen L. Olivier, an
 d Andrew J. Younge (Sandia National Laboratories)\n---------------------\n
 Containerization in HPC Environments: Challenges in Optimizing Performance
  and Security in Scientific Computing\n\nThe convergence of scientific com
 puting, exascale data, and high-performance computing (HPC) has driven sig
 nificant advancements across multiple scientific domains. However, the gro
 wing complexity of software stacks, coupled with the increased need for re
 producibility and efficient resource manageme...\n\n\nVijayalakshmi Sarava
 nan (University of Texas at Tyler, Brookhaven National Laboratory) and Kha
 led Z. Ibrahim (Lawrence Berkeley National Laboratory (LBNL))\n-----------
 ----------\nEnabling HPC Scientific Workflows for Serverless\n\nThe conver
 gence of edge computing, big data analytics, and AI with traditional scien
 tific calculations is increasingly being adopted in HPC workflows. Workflo
 w management systems are crucial for managing and orchestrating these comp
 lex computational tasks. However, it is difficult to identify patter...\n\
 n\nAnderson Andrei Da Silva, Rolando Pablo Hong Enriquez, Gourav Rattihall
 i, and Vijay Thurimella (Hewlett Packard Labs); Rafael Ferreira Da Silva (
 Oak Ridge National Laboratory (ORNL)); and Dejan Milojicic (Hewlett Packar
 d Labs)\n---------------------\nPULSE: Using Mixed-Quality Models for Redu
 cing Serverless Keep-Alive Cost\n\nThis paper addresses a key challenge wi
 th using serverless computing for machine learning (ML) inference, which i
 s cold starts that occur during initial invocations and container inactivi
 ty. Fixed keep-alive policies, like the commonly adopted 10-minute strateg
 y, have been implemented by cloud provi...\n\n\nKausalya Sankaranarayanan,
  Rohan Basu Roy, and Devesh Tiwari (Northeastern University)\n------------
 ---------\nCANOPIE-HPC Community Discussion/Open Q&A\n\nAlberto Madonna (S
 wiss National Supercomputing Centre (CSCS)) and Claudia Misale (IBM TJ Wat
 son Research Center)\n---------------------\nDeploying Containers on Secur
 e HPC Systems\n\nThere is increasing interest in the use of containerized 
 workflows on high performance computing (HPC) systems from the research co
 mmunity and industry to tackle “real world” problems, which requires HPC r
 esources to efficiently compute and scale complex algorithms across thousa
 nds of no...\n\n\nDavid Brayford (Hewlett Packard Labs)\n-----------------
 ----\nXaaS: Acceleration as a Service to Enable Productive High-Performanc
 e Cloud Computing\n\nHPC and cloud computing have historically evolved ind
 ependently, each specializing in performance or productivity. But what if 
 we could combine the best of both worlds? The recently proposed Accelerati
 on as a Service (XaaS) presented a new vision for the future — a unified a
 rchitecture providi...\n\n\nMarcin Copik and Torsten Hoefler (ETH Zürich),
  Thomas Schulthess (Swiss National Supercomputing Centre (CSCS)), Dan Erns
 t (NVIDIA Corporation), and Ian Foster (Argonne National Laboratory)\n----
 -----------------\nAfternoon Break\n---------------------\nContainerized C
 heckpoint-Restart Mechanisms for HPC\n\nHigh-performance computing (HPC) s
 ystems are crucial for solving complex scientific problems but challenges 
 like resource management, fault tolerance, and maintaining consistent perf
 ormance across diverse environments can be difficult. Container technologi
 es, like NERSC's Shifter and Podman-HPC, off...\n\n\nMadan Timalsina and N
 icholas Tyler (NERSC at LBNL (Lawrence Berkeley National Laboratory))\n---
 ------------------\nApplying Container Technology to the NOAA-GFDL MSD FRE
  Workflow, Part 1\n\nThe Flexible Modeling System (FMS) Runtime Environmen
 t (FRE) handles post processing of the National Atmospheric and Oceanic Ad
 ministration (NOAA) research climate models on a computing cluster called 
 Post Processing and Analysis (PPAN). FRE is currently only used on PPAN, b
 ut NOAA wishes to provide...\n\n\nCiheim Brown, Dana Singh, and Thomas Rob
 inson (Geophysical Fluid Dynamics Laboratory (NOAA))\n--------------------
 -\nCANOPIE-HPC — Introduction and Welcome\n\nAlberto Madonna (Swiss Nation
 al Supercomputing Centre (CSCS)), Claudia Misale (IBM TJ Watson Research C
 enter), Shane Canon (Lawrence Berkeley National Laboratory (LBNL)), and An
 drew Younge (Sandia National Laboratories)\n---------------------\nLunch B
 reak\n\nTag: Cloud Computing, Middleware and System Software, State of the
  Practice\n\nRegistration Category: Workshop Reg Pass\n\nSession Chairs: R
 ichard Shane Canon (Lawrence Berkeley National Laboratory (LBNL)); Alberto
  Madonna (ETH Zürich, Swiss National Supercomputing Centre (CSCS)); Claudi
 a Misale (CoreWeave, Inc.); and Andrew Younge (Sandia National Laboratorie
 s)
END:VEVENT
END:VCALENDAR