OpenStack-HPC

Download

openstack-hpc@lists.openstack.org

April 2024

1 participants
2 discussions

[CALL FOR PAPER] RT-Cloud 2024 - The third International Workshop on Real-Time Cloud Systems
by Remo Andreoli 21 Apr '24

21 Apr '24

-------------------------------------------------------------------------------- The third International Workshop on Real-Time Cloud Systems (RT-Cloud 2024) www.ecrts.org/rt-cloud-2024/ -------------------------------------------------------------------------------- While most cloud services operate on a best-effort basis and provide no timing guarantees, there is a growing interest in cloud technologies for industrial and critical applications. Edge computing is a recent paradigm that brings processing and storage resources closer to the user. As a result, it provides benefits such as low latency and high bandwidth while still providing the benefits of cloud computing, making it lucrative for low-latency, real-time / safety-critical applications. However, a higher degree of determinism is desired for cloud and edge technologies to be fully embraced by the real-time / safety-critical industry. This calls for new techniques and methodologies to design predictable and reliable cloud/edge technologies and applications. The 3rd International Workshop on Real-Time Cloud Systems (RT-Cloud) aims to bring together industrial and academic researchers and become a forum for discussing topics (including ideas, open and upcoming challenges, use cases, initiatives, calls to action, and projects, regulations and standards, future research directions, etc.) related to real-time cloud/ edge/fog computing and virtualization techniques. This year, the workshop will have a focus on industrial cloud use cases. The workshop also welcomes short and full papers that consider: - New technologies, innovative approaches/ideas, including work-in-progress - Open problems and upcoming challenges - Regulations and standards - Demos and tutorials, especially, for open-source projects - Call to actions - Experimental reports on existing technologies, including negative ones Original full and short papers on all aspects of real-time/safety-critical cloud/edge/fog and virtualization are welcome. Researchers are encouraged (but not limited) to submit papers regarding the following: 1. Use cases and applications for RT-cloud - Use of cloud computing principles to improve performance/safety/ security in existing real-time systems/use cases - Real-time cloud and edge applications -- Industrial use cases and requirements -- Experiments and experiences with RT-cloud -- Real-time / safety-critical applications as a cloud-based service -- Real-time extension to existing cloud-computing technologies -- Decomposition between the edge, fog, and cloud 2. Virtualization and cloud computing technologies/models for RT-cloud - Development Operations (DevOps) for RT-cloud - Deterministic hypervisors and cloud execution environments, including containers and WASM - Real-time Saas, Paas, Iaas, and XaaS (anything as a service model) - Execution and deployment models for RT-Cloud, including serverless computing - Microservices Architectures for real-time system - Modeling of network and cloud, execution stacks, and applications - Adaptive SLAs for RT-cloud 3. Resource management (including monitoring, scheduling) and orchestration for RT-cloud -- Predictable orchestration and cloud operation -- Tools and techniques for resource sharing and isolation -- End-to-end resource management and scheduling in edge-cloud continuum -- Co-scheduling of virtual network and compute resources -- Edge-cloud-IoT continuum and interaction/interfacing -- QoS mechanisms and isolation guarantees -- Autonomous monitoring systems 4. Safety and Security for RT-cloud - Assurance in RT-cloud, deterministic scaling, fail-over, migration, etc. - Trusted RT-Cloud/edge environment -- Considering security and privacy together with safety -------------------------------------------------------------------------------- IMPORTANT DATES Submission deadline: May 9, 2024 (AoE) Notification of Acceptance: June 6, 2024 RT-Cloud Workshop: July 9, 2024 ECRTS 2024: July 9-12, 2024 -------------------------------------------------------------------------------- SUBMISSION INSTRUCTIONS Manuscripts must present original work. They must be written in English, using the IEEE format, and submitted through easychair. There are no strict rules about the number of pages, but the recommended length for full papers is 6 to 8 pages, and for short papers is 3-4 pages. Submission link: https://easychair.org/conferences/?conf=rtcloud2024 All submissions will be reviewed by the technical program committee members, and accepted papers will be included in the workshop proceedings, which will be available online on the workshop website. Papers will not be assigned a DOI to enable authors to submit a full/extended paper later on to a conference or a journal. For every accepted paper, at least one author is expected to register for the workshop and present the paper. -------------------------------------------------------------------------------- WORKSHOP CHAIRS - Johan Eker (Lund University/Ericsson) - Luca Abeni (Scuola Superiore Sant'Anna - luca.abeni(a)santannapisa.it) - Gautam Gala (University of Kaiserslautern-Landau - gala(a)eit.uni-kl.de) PUBLICITY CHAIR - Remo Andreoli (Scuola Superiore Sant'Anna - remo.andreoli(a)santannapisa.it)

1 0

VHPC'24 - 19th Virtualization in High-Performance Cloud Computing Workshop - Call for Papers
by Remo Andreoli 07 Apr '24

07 Apr '24

VHPC'24<https://vhpc.org>: August 26th-27th, Madrid, Spain, co-located with Euro-Par 2024. Papers due: May 6th. The Workshop on Virtualization in High-Performance Cloud Computing (VHPC<https://vhpc.org>) aims to bring together researchers and industrial practitioners facing the challenges posed by virtualization in HPC/Cloud scenarios in order to foster discussion, collaboration, mutual exchange of knowledge and experience, enabling research to ultimately provide novel solutions for virtualized computing systems. The workshop features paper presentations, discussion sessions, and lightning talks. Accepted papers will be published in a Springer LNCS volume (indexed by Scopus and others). Focus Topics This year, we are calling the timely topic of virtualization in support of high-memory LLM training workloads including, but not limited to: * GPU hypervisor memory virtualization: Techniques for virtualizing GPU memory to allow flexible and efficient allocation across multiple workloads, enabling higher utilization of GPU resources * Flat CPU/GPU memory page tables/TLB: Unified virtual memory spaces and page table structures that allow both GPUs to address CPU memory mapped to accelerator global memory space * Storage/filesystem to virtual memory mapped approaches * Distributed memory virtualization * Memory compression and reduction techniques: approaches for compressing model parameters, activations, and gradients to reduce memory requirements during training * Out-of-core training algorithms * Efficient memory allocation and management: Techniques for optimizing memory allocation, reducing fragmentation, and improving memory utilization during training * Memory-efficient data formats and processing: Data formats and processing techniques that minimize memory overhead while maintaining training efficiency * Benchmarking and profiling tools: Tools and methodologies for measuring, analyzing, and optimizing memory usage in LLM training workloads * Case studies and applications: Real-world examples and applications of virtualization techniques in LLM training scenarios In addition to regular topics: * Container Platforms (Kubernetes, Docker, Nitro/Firecracker, Singularity, Shifter, rkt, …) and Unikernel Frameworks * Composable Lightweight Application Components and Lambda / Function-as-a-Service Paradigms * Latency Control, Warm/Cold-start issues and Data/Container Placement in Heterogeneous HPC Virtualized Environments * Energy-efficiency and Service Orchestration in Virtualized Cloud & HPC Infrastructures The call for papers is also available in PDF Format<https://vhpc.github.io/VHPC-CfP.pdf>. The workshop will be composed of 20 min paper presentations, each followed by 10 min discussion sessions, plus lightning talks that are limited to 5 minutes. Presentations may be accompanied by interactive demonstrations. For more information and detailed paper submission instructions, refer to the VHPC'24 webpage<https://vhpc.org/>. Important Dates (AoE) * May 6th, 2024: Paper submission deadline (Springer LNCS) * Jun 20th, 2024: Acceptance notification * Jul 1st, 2024: Camera-ready version due * Aug 26th-27th, 2024: Workshop Day General Chairs * Michael Alexander, Institute of Science and Technology, Austria * Anastassios Nanos, Nubificus Ltd., UK * Tommaso Cucinotta, Scuola Superiore Sant'Anna, Italy Publicity Chairs * Remo Andreoli, Scuola Superiore Sant'Anna, Italy General Information The workshop will be held in conjuction with the International European Conference on Parallel and Distributed Computing (Euro-Par 2024)<https://2024.euro-par.org>, on Aug 26th-27th, 2024, in Madrid, Spain. The VHPC co-chairs

1 0