Ceph的动态

?Cephalocon 2024 has kicked off at CERN? It was both an honour and a privilege to open the Cephalocon 2024 hosted at the iconic #CERNGLOBE. This week, over 350 participants unite not only as developers, operators, and researchers but as a global community brought together by a passion for #Ceph?- the revolutionary open-source distributed storage system. Ceph is more than a technology; it is a testament to the power of #collaboration and #innovation. With its flexibility, scalability, and fault-tolerant architecture, Ceph continues to reshape how we think about and manage data. From cloud-native environments to research giants like CERN, Ceph allows us to navigate the exponential growth of data in ways that are both efficient and adaptable. It’s fitting that Cephalocon 2024 is hosted at CERN, an institution that represents the very spirit of?open science. Just as CERN thrives on transparency, shared learning, and collaboration, Ceph too flourishes in an open-source environment driven by a community of passionate contributors. Cephalocon is a celebration of what we can achieve when we come together—not in silos, but as a collective striving toward a common goal. With a 30% increase in registrations from last year, it's clear that our community is growing and thriving. A heartfelt thank you to the Linux Foundation and the CEPH foundation for co-hosting, the organizing committee, and all who have contributed to making this community gather a reality!…connect, learn, and innovate….Together, we are shaping the future of data storage. ?#Cephalocon2024 #Ceph #OpenSource #Innovation #Collaboration #CERN #DataStorage #TechCommunity #DistributedSystems #OpenScience?@cerncomputing https://lnkd.in/d-WcYqGT

?? Only at #Cephalocon ?? ?? 20 Years of Ceph Dan van der Ster of CLYSO will discuss why Ceph remains the leading software-defined storage solution. This talk will cover its evolution since 2004, highlighting innovations in unified block, object, and file storage. Learn about CRUSH, BlueStore, and stretch clusters that enhance scalability and performance, especially for AI workloads. More info: https://lnkd.in/e3GvrfJc #ceph #cephalocon #opensource #storage

Interested in using remote workflows at ALCF? ? In this video, Argonne National Laboratory's Christine Simpson PhD demonstrates how to use Globus.org tools—Globus Compute and Globus Flows—to integrate the ALCF's HPC systems with distributed workloads involving data movement and remote orchestration. ? Watch here: https://lnkd.in/gzNaW_hU ? To view slides from the session, visit: https://lnkd.in/gpDCuNaN

"You want to be able to have users sending requests from outside Aurora to get inference results back from a service like AuroraGPT..." "...We have an increasing number of users that have workloads that need to be triggered in this way" (e.g. from their laptop or a system at a remote facility). This talk provides a useful deep dive with concrete examples and documentation on a range of use cases for remote workload management. #hpc #ai #globus #alcf

Here is a great use case featuring Globus Compute and Globus Flows. Watch the video! #researchdatamanagement #dataautomation

I am very proud to say our paper "Falcon: Live Reconfiguration for Stateful Stream Processing on the Edge", has just won the ?? Best Paper Award ?? at the The Ninth ACM/IEEE Symposium on Edge Computing (SEC' 24). The Falcon distributed stream processing framework provides seamless reconfiguration of stateful operators, with practically no disruption to the running application, even across distant datacenters. It reduces the length of processing interruptions and their impact on latency by 2 to 4 orders of magnitude, compared to state-of-the-art frameworks such as Flink, Trisk, and Meces. The prototype (and soon link to paper) is available at https://lnkd.in/eEiH7Th3 This amazing achievement was only possible thanks to the hard work of main author Pritish Mishra and amazing co-authors Nelson Bore, Brian Ramprasad, Myles Thiessen, Alexandre da Silva Veith, Oana Balmau, and Eyal de Lara! #EdgeComputing #EdgeAI #Systems

SuperComputing (SC) is the main event for High Performance Computing, Networking, Storage and Analysis. I am very proud of Rapha?l Frantz and Jerónimo Sánchez García, which both landed first authored papers there within the first year of their PhD in the framework of the QUARC Project. This speaks highly of them as researchers and the team behind supporting their developments. They are, as we speak, ready to roll in Atlanta. This is the outcome of a great private-public partnership - with joint efforts from Idelfonso Tafur Monroy from Eindhoven University of Technology, Sokol Kosta from Aalborg University, Marcin Copik and Torsten Hoefler ???? from ETH Zürich, and Salvatore Di Girolamo, Juan Jose (JJ) Vegas Olmos, Rami Nudelman, Gil Bloch from NVIDIA, in the context of the Doctoral Network QUARC Project. You can find more info on the papers here: Protocol Buffer Deserialization DPU offloading in the RPC Datapath, by Rapha?l Frantz https://lnkd.in/dae8NVhh Offloaded MPW Message Matching: An Optimistic Approach, by Jerónimo Sánchez García https://lnkd.in/dcG8FYgR #sc24 #industrialPhD

Quantum technologies are evolving rapidly, introducing new capabilities that are transforming how Federal agencies approach innovation and data protection. As quantum computing, sensing and security advance, agencies must stay ahead by understanding their implications and preparing for quantum-safe transitions. On February 27, join Carahsoft, Quintessence Labs and Bob Gourley, CTO at OODA LLC, for an in-depth discussion on the latest breakthroughs in quantum technology and evolving Federal policies shaping quantum research and adoption. Attendees will gain insights into encryption strategies, quantum key distribution and strategies for achieving crypto-agility. Sign up today to stay informed on the future of quantum innovation and security: https://ow.ly/wHk130sJFgk #QuantumComputing #QuantumSecurity #FederalIT #QuantumInnovation

UnitaryHack is running! You can join at any moment. The format is super simple: fix a GitHub issue of some quantum computing software library and get the associated bounty.

Are you trying to move to the next level of quantum computing? Fault tolerant quantum computing requires scalable quantum error correction. The Qblox Cluster's modular architecture seeamlessly scales to provide precise control and efficient readout of individual and coupled qubits, facilitating the study of their interactions. Learn more about fully parallelized operations, perfect timing control, and how the #qblox system can help you: https://hubs.ly/Q02NjZnM0