Conveners
Scalable Storage Backends and Integration with Data Processing
- Maciej Brzezniak
Scalable Storage Backends and Integration with Data Processing
- Maciej Brzezniak
-
Huseyn Gurbanov (Leibniz Supercomputing Centre (LRZ) of the BAdW)19/03/2026, 14:00Scalable Storage Backends and Integration with Data Processing
Amazon's Simple Storage Service (S3) protocol has become a de-facto standard for object storage, and academic supercomputing centres are increasingly interested in offering S3-compatible storage solutions. This is on the one hand due to more (on-premises) cloud computing at such centres, and on the other hand due to the increasing demands of flexible data access and sharing across computing...
Go to contribution page -
Adam Prycki19/03/2026, 14:15EFSS and storage federations for EOSC and eResearch infrastructures.
Title: Feasibility study trans-national replicated object storage using Ceph
Subtitle: Testing key technical components of a potential resilient pan-European object storage infrastructureAbstract: The European NREN community is working on a proposal for a large scale sovereign object storage infrastructure on European scale. To test key technical aspects, four NRENs are currently...
Go to contribution page -
Krzysztof Wadówka (PSNC)19/03/2026, 14:30Scalable Storage Backends and Integration with Data Processing
S3 protocol is the de-facto standard data protocol in distributed storage and cloud computing with its wide usage for cloud back-ends, back-up and long-term storage. It is also being adopted in HPC computing systems and AI platforms. This trend is seen in academia and is strengthening in industry with e.g. nVidia assuming S3 to be the default data protocol for accessing data from GPUs for AI...
Go to contribution page -
Mario Vitale (CERN)19/03/2026, 14:45Scalable Storage Backends and Integration with Data Processing
The CERN Tape Archive (CTA) is the open source solution developed at CERN to store more than 1 Exabyte of data from CERN’s experimental programmes. CTA interfaces with two disk systems widely used by the High-Energy Physics (HEP) community, EOS and dCache. However, until now there has been no integration with systems used outside of HEP.
Looking at current industry standards, the leading...
Go to contribution page -
Mr Piotr Modrzyk (Leil Storage)19/03/2026, 16:00Scalable Storage Backends and Integration with Data Processing
As HDD capacities reach tens of terabytes while per drive IOPS remain largely unchanged, conventional SDS designs often hit IOPS limits before achieving full sequential bandwidth. This imbalance frequently leads operators to choose smaller disks in an attempt to balance capacity and performance. Leil Storage addresses this challenge with a sequential access model that aligns I/O patterns with...
Go to contribution page -
Ingo Ebel (Deutsches Elektronen-Synchrotron (DE))19/03/2026, 16:15Scalable Storage Backends and Integration with Data Processing
DESY completed a migration of user home directories from AFS to CephFS to address performance limitations, aging infrastructure, and operational complexity. The project involved deploying a high-availability Ceph cluster with NVME-backed metadata pools, erasure-coded data pools and quota management - demonstrating CephFS as a scalable, production-grade replacement for legacy AFS deployments in...
Go to contribution page -
Dr Jörn Dreyer (OpenCloud GmbH)19/03/2026, 16:30File Sync & Share Solutions and Requirement from the Community
OpenCloud has the design goal to not use a relational database. This requires a deeper integration with the underlying storage system, ie. through extensive use of extended file attributes. Since features like file revisions, trash and shares are inevitable nowadays, OpenCloud makes use of SDS native supported storage aspects to build these advanced features...
Go to contribution page -
kritik sachdeva (IBM)19/03/2026, 16:45Scalable Storage Backends and Integration with Data Processing
Operating Ceph RADOS Gateway (RGW) as a backend for sync & share services presents unique challenges that differ from typical S3 workloads. Drawing from hands-on experience supporting RGW deployments across diverse production environments, this presentation shares practical operational insights that service operators can immediately apply.
We examine the most common issues encountered in...
Go to contribution page