@techreport{62981, abstract = {{Otus is a high-performance computing cluster that was launched in 2025 and is operated by the Paderborn Center for Parallel Computing (PC2) at Paderborn University in Germany. The system is part of the National High Performance Computing (NHR) initiative. Otus complements the previous supercomputer Noctua 2, offering approximately twice the computing power while retaining the three node types that were characteristic of Noctua 2: 1) CPU compute nodes with different memory capacities, 2) high-end GPU nodes, and 3) HPC-grade FPGA nodes. On the Top500 list, which ranks the 500 most powerful supercomputers in the world, Otus is in position 164 with the CPU partition and in position 255 with the GPU partition (June 2025). On the Green500 list, ranking the 500 most energy-efficient supercomputers in the world, Otus is in position 5 with the GPU partition (June 2025). This article provides a comprehensive overview of the system in terms of its hardware, software, system integration, and its overall integration into the data center building to ensure energy-efficient operation. The article aims to provide unique insights for scientists using the system and for other centers operating HPC clusters. The article will be continuously updated to reflect the latest system setup and measurements. }}, author = {{Ehtesabi, Sadaf and Hossain, Manoar and Kenter, Tobias and Krawinkel, Andreas and Ostermann, Lukas and Plessl, Christian and Riebler, Heinrich and Rohde, Stefan and Schade, Robert and Schwarz, Michael and Simon, Jens and Winnwa, Nils and Wiens, Alex and Wu, Xin}}, keywords = {{Otus, Supercomputer, FPGA, PC2, Paderborn Center for Parallel Computing, Noctua 2, HPC}}, pages = {{33}}, publisher = {{Paderborn Center for Parallel Computing (PC2)}}, title = {{{Otus Supercomputer}}}, doi = {{10.48550/ARXIV.2512.07401}}, volume = {{1}}, year = {{2025}}, } @article{53663, abstract = {{Noctua 2 is a supercomputer operated at the Paderborn Center for Parallel Computing (PC2) at Paderborn University in Germany. Noctua 2 was inaugurated in 2022 and is an Atos BullSequana XH2000 system. It consists mainly of three node types: 1) CPU Compute nodes with AMD EPYC processors in different main memory configurations, 2) GPU nodes with NVIDIA A100 GPUs, and 3) FPGA nodes with Xilinx Alveo U280 and Intel Stratix 10 FPGA cards. While CPUs and GPUs are known off-the-shelf components in HPC systems, the operation of a large number of FPGA cards from different vendors and a dedicated FPGA-to-FPGA network are unique characteristics of Noctua 2. This paper describes in detail the overall setup of Noctua 2 and gives insights into the operation of the cluster from a hardware, software and facility perspective.}}, author = {{Bauer, Carsten and Kenter, Tobias and Lass, Michael and Mazur, Lukas and Meyer, Marius and Nitsche, Holger and Riebler, Heinrich and Schade, Robert and Schwarz, Michael and Winnwa, Nils and Wiens, Alex and Wu, Xin and Plessl, Christian and Simon, Jens}}, journal = {{Journal of large-scale research facilities}}, keywords = {{Noctua 2, Supercomputer, FPGA, PC2, Paderborn Center for Parallel Computing}}, title = {{{Noctua 2 Supercomputer}}}, doi = {{10.17815/jlsrf-8-187 }}, volume = {{9}}, year = {{2024}}, } @article{1965, abstract = {{Virtualization technology makes data centers more dynamic and easier to administrate. Today, cloud providers offer customers access to complex applications running on virtualized hardware. Nevertheless, big virtualized data centers become stochastic environments and the simplification on the user side leads to many challenges for the provider. He has to find cost-efficient configurations and has to deal with dynamic environments to ensure service level objectives (SLOs). We introduce a software solution that reduces the degree of human intervention to manage clouds. It is designed as a multi-agent system (MAS) and placed on top of the Infrastructure as a Service (IaaS) layer. Worker agents allocate resources, configure applications, check the feasibility of requests, and generate cost estimates. They are equipped with application specific knowledge allowing it to estimate the type and number of necessary resources. During runtime, a worker agent monitors the job and adapts its resources to ensure the specified quality of service—even in noisy clouds where the job instances are influenced by other jobs. They interact with a scheduler agent, which takes care of limited resources and does a cost-aware scheduling by assigning jobs to times with low costs. The whole architecture is self-optimizing and able to use public or private clouds. Building a private cloud needs to face the challenge to find a mapping of virtual machines (VMs) to hosts. We present a rule-based mapping algorithm for VMs. It offers an interface where policies can be defined and combined in a generic way. The algorithm performs the initial mapping at request time as well as a remapping during runtime. It deals with policy and infrastructure changes. An energy-aware scheduler and the availability of cheap resources provided by a spot market are analyzed. We evaluated our approach by building up an SaaS stack, which assigns resources in consideration of an energy function and that ensures SLOs of two different applications, a brokerage system and a high-performance computing software. Experiments were done on a real cloud system and by simulations.}}, author = {{Niehörster, Oliver and Simon, Jens and Brinkmann, André and Keller, Axel and Krüger, Jens}}, journal = {{Journal of Grid Computing}}, number = {{3}}, pages = {{553--577}}, title = {{{Cost-aware and SLO Fulfilling Software as a Service}}}, doi = {{10.1007/s10723-012-9230-7}}, volume = {{10}}, year = {{2012}}, } @article{1971, abstract = {{System virtualization has become the enabling technology to manage the increasing number of different applications inside data centers. The abstraction from the underlying hardware and the provision of multiple virtual machines (VM) on a single physical server have led to a consolidation and more efficient usage of physical servers. The abstraction from the hardware also eases the provision of applications on different data centers, as applied in several cloud computing environments. In this case, the application need not adapt to the environment of the cloud computing provider, but can travel around with its own VM image, including its own operating system and libraries. System virtualization and cloud computing could also be very attractive in the context of high‐performance computing (HPC). Today, HPC centers have to cope with both, the management of the infrastructure and also the applications. Virtualization technology would enable these centers to focus on the infrastructure, while the users, collaborating inside their virtual organizations (VOs), would be able to provide the software. Nevertheless, there seems to be a contradiction between HPC and cloud computing, as there are very few successful approaches to virtualize HPC centers. This work discusses the underlying reasons, including the management and performance, and presents solutions to overcome the contradiction, including a set of new libraries. The viability of the presented approach is shown based on evaluating a selected parallel, scientific application in a virtualized HPC environment. }}, author = {{Birkenheuer, Georg and Brinkmann, André and Kaiser, Jürgen and Keller, Axel and Keller, Matthias and Kleineweber, Christoph and Konersmann, Christoph and Niehörster, Oliver and Schäfer, Thorsten and Simon, Jens and Wilhelm, Maximilan}}, journal = {{Software: Practice and Experience}}, publisher = {{John Wiley & Sons}}, title = {{{Virtualized HPC: a contradiction in terms?}}}, doi = {{10.1002/spe.1055}}, year = {{2011}}, } @inproceedings{2203, author = {{Niehörster, Oliver and Simon, Jens and Brinkmann, André and Krieger, Alexaner}}, booktitle = {{Proc. IEEE/ACM Int. Conf. on Grid Computing (GRID)}}, isbn = {{978-0-7695-4572-1}}, pages = {{157--164}}, publisher = {{IEEE Computer Society}}, title = {{{Autonomic Resource Management with Support Vector Machines}}}, doi = {{10.1109/Grid.2011.28}}, year = {{2011}}, } @inproceedings{2237, author = {{Niehörster, Oliver and Brinkmann, André and Fels, Gregor and Krüger, Jens and Simon, Jens}}, booktitle = {{Proc. Int. Conf. on Cluster Computing (CLUSTER)}}, issn = {{1552-5244}}, pages = {{178--187}}, publisher = {{IEEE}}, title = {{{Enforcing SLAs in Scientific Clouds}}}, doi = {{10.1109/CLUSTER.2010.42}}, year = {{2010}}, } @inproceedings{2357, author = {{Birkenheuer, Georg and Brinkmann, André and Dömer, Hubert and Effert, Sascha and Konersmann, Christoph and Niehörster, Oliver and Simon, Jens}}, booktitle = {{Proc. Gemeinsamer Workshop der GI/ITG Fachgruppen "Betriebssysteme" und "KuVS": Virtualized IT infrastructures and their management}}, pages = {{37--49}}, publisher = {{Leibniz-Rechenzentrum}}, title = {{{Virtual Supercomputer for HPC and HTC}}}, year = {{2008}}, } @inproceedings{2426, author = {{P. Miller, Barton and Labarta, Jesús and Schintke, Florian and Simon, Jens}}, booktitle = {{Proc. European Conf. on Parallel Processing (Euro-Par)}}, isbn = {{978-3-540-45706-0}}, pages = {{131}}, publisher = {{Springer}}, title = {{{Performance Evaluation, Analysis and Optimization}}}, doi = {{10.1007/3-540-45706-2_15}}, volume = {{2400}}, year = {{2002}}, } @inproceedings{2431, author = {{Schintke, Florian and Simon, Jens and Reinefeld, Alexander}}, booktitle = {{Proc. Int. Conf. on Computational Science (ICCS)}}, pages = {{569--578}}, publisher = {{Springer}}, title = {{{A Cache Simulator for Shared Memory Systems}}}, doi = {{10.1007/3-540-45718-6_62}}, volume = {{2074}}, year = {{2001}}, } @phdthesis{2434, author = {{Simon, Jens}}, isbn = {{3-934445-03-9}}, pages = {{255}}, publisher = {{Dr. Dirk Blunk Verlag}}, title = {{{Werkzeugunterstützte effiziente Nutzung von Hochleistungsrechnern}}}, year = {{2000}}, } @inbook{2435, author = {{Simon, Jens and Reinefeld, Alexander and Heinz, Oliver}}, booktitle = {{SCI: Scalable Coherent Interface. Architecture and Software for High-Performance Compute Clusters}}, editor = {{Hellwagner, Hermann and Reinefeld, Alexander}}, isbn = {{978-3-540-47048-9}}, issn = {{0302-9743}}, pages = {{367--381}}, publisher = {{Springer}}, title = {{{Large-Scale SCI Clusters in Practice: Architecture and Performance in SCI}}}, doi = {{10.1007/10704208}}, volume = {{1734}}, year = {{1999}}, } @article{2437, author = {{Simon, Jens and Wierum, Jens-Michael}}, issn = {{0020-0190}}, journal = {{Information Processing Letters - Special Issue on Models of Computation}}, number = {{5}}, pages = {{255--261}}, publisher = {{Elsevier}}, title = {{{The Latency-of-Data-Access model for Analyzing Parallel Computation}}}, doi = {{10.1016/S0020-0190(98)00062-3}}, volume = {{66}}, year = {{1998}}, } @inproceedings{2438, author = {{Simon, Jens and Weicker, Reinhold and Vieth, Marco}}, booktitle = {{Proc. European Conf. on Parallel Processing (Euro-Par)}}, isbn = {{978-3-540-69549-3}}, pages = {{971--984}}, publisher = {{Springer}}, title = {{{Workload Analysis of Computation Intensive Tasks: Case Study on SPEC CPU95 Benchmarks}}}, doi = {{10.1007/BFb0002841}}, volume = {{1300}}, year = {{1997}}, } @inproceedings{2439, author = {{Heinz, Oliver and Simon, Jens}}, booktitle = {{Proc. Int. Conf. on Architecture of Computing Systems (ARCS)}}, publisher = {{VDE Verlag}}, title = {{{Experiences with a SCI Multiprocessor Workstation Cluster}}}, year = {{1997}}, } @inproceedings{2440, author = {{Simon, Jens and Heinz, Oliver}}, booktitle = {{Proc. Workshops im Rahmen der 14. ITG/GI-Fachtagung Architektur von Rechensystemen}}, pages = {{189--199}}, title = {{{SCI multiprocessor PC cluster in a WindowsNT environment}}}, year = {{1997}}, } @inproceedings{2441, author = {{Fischer, Markus and Simon, Jens}}, booktitle = {{Proc. European Parallel Virtual Machine / Message Passing Interface Users’ Group Meeting (EuroPVM/MPI)}}, pages = {{175--184}}, publisher = {{Springer}}, title = {{{Embedding SCI into PVM}}}, doi = {{10.1007/3-540-63697-8_84}}, volume = {{1332}}, year = {{1997}}, } @inproceedings{2442, author = {{Reinefeld, Alexander and Baraglia, Ranieri and Decker, Thomas and Gehring, Jörn and Laforenza, Domenico and Ramme, Friedhelm and Römke, Thomas and Simon, Jens}}, booktitle = {{Proc. Heterogenous Computing Workshop (HCW)}}, pages = {{17--31}}, publisher = {{IEEE Computer Society}}, title = {{{The MOL Project: An Open, Extensible Metacomputer}}}, doi = {{10.1109/HCW.1997.581407}}, year = {{1997}}, } @inproceedings{2443, author = {{Simon, Jens and Wierum, Jens-Michael}}, booktitle = {{Proc. Int. Conf. on High-Performance Computing and Networking (HPCN-Europe)}}, isbn = {{978-3-540-61142-4}}, pages = {{627--632}}, publisher = {{Springer}}, title = {{{Sequential Performance versus Scalability: Optimizing Parallel LU-Decomposition}}}, doi = {{10.1007/3-540-61142-8_606}}, volume = {{1067}}, year = {{1996}}, } @inproceedings{2444, author = {{Simon, Jens and Wierum, Jens-Michael}}, booktitle = {{Proc. Annual Int. Conf. on High-Performance Computers (HPCS)}}, title = {{{Performance Prediction of Benchmark Programs for Massively Parallel Architectures}}}, year = {{1996}}, } @inproceedings{2445, author = {{Simon, Jens and Wierum, Jens-Michael}}, booktitle = {{Proc. European Conf. on Parallel Processing (Euro-Par)}}, pages = {{675--688}}, publisher = {{Springer}}, title = {{{Accurate Performance Prediction for Massively Parallel Systems and its Applications}}}, doi = {{10.1007/BFb0024764}}, volume = {{1124}}, year = {{1996}}, }