Industrieseminar
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Datum | Name | Organisation | Informationen |
|---|---|---|---|
| 28.10.2010 | Keine Veranstaltung | ||
| 04.11.2010 | Walfried Wagener | Biotronik |
Cloud Computing@BIOTRONIK (Folien) |
| 11.11.2010 | Dr. Alexander Gebhart | Director Virtualization and Cloud Computing, Technology and Innovation Platform, SAP AG |
Virtualization and Cloud Management (Folien)Inhalt des Vortrages ist das Management von SAP Systemen im Kontext von Virtualisierung und Cloud Computing. Die Abteilung "Virtualization and Cloud Management" ist ein Teil des Bereiches "Technology and Innovation Platform (TIP)". |
| 18.11.2010 | Keine Veranstaltung | ||
| 25.11.2010 | Andreas Giesekus | GKVI |
Virtualisierung und Konsolidierung von Unix-Plattformen bei der gkv informatik (Folien)Die gkv informatik betreibt die IT-Systeme für fünf AOK-Kunden und die BARMER GEK. Ein Kernstück der Anwendungslandschaft ist eine SAP-basierte Branchenlösung für die gesetzlichen Krankenversicherungen. UNIX-Plattformen spielen in diesem Zusammenhang eine zunehmend wichtige Rolle. Gegenstand des Vortrags ist ein Bericht zur Architektur der installierten Serverlandschaft in der gkvi informatik und ihrer strategischen Ausrichtung. |
| 02.12.2010 | Dr. Frank Baetke | Hewlett-Packard ISS/SCI |
Implementations of Standards-based and Energy-efficient High-end Cloud-resourceMost high-end systems that will end-up in Cloud- (and Grid-) Environments will not only be competing for users in terms of performance but also in terms of operational cost. Certain aspects will significantly gain in importance in the years to come.Beyond acquisition cost, a major factor is power and cooling efficiency. This is primarily an issue of cost for power, but also for the power and thermal density of what can be managed in a data center. To leverage the economics of scale established HPC centers as well as providers of innovative services are evaluating new concepts which have the potential to make classical data center designs obsolete. Those new concepts provide significant advantages in terms of energy efficiency, deployment flexibility and manageability. Examples of this new approach, often dubbed POD for Performance Optimized Datacenter, including a concept to scale to multiple PFLOPS at highest energy efficiency will be shown. |
| 09.12.2010 | Dr. Jacob Gorm Hansen | VMWare |
Virtualization as a foundation for the CloudBecause they provide a well-defined and safe sandboxing environment with near-native performance, virtual machines have become the fundamental building block of cloud computing. Virtualization hides most physical details about the computer on which a workload is running, enabling live migration of data and VMs, to balance load within a data center, and to allow customers to quickly switch providers in search of better or cheaper services. In this talk, I will describe the research that led to NomadBIOS, the first hypervisor to support live VM migration, and the follow-up work that spawned both Xen's live migration, and the novel and light-weight "self-migration" prototype that allowed VMs to migrate without any hypervisor assistance, for improved security and flexibility. Looking forward, I will describe my recent work at VMware on "Lithium", a scalable and high-performance storage system for VMs in the cloud, and discuss some future challenges of virtualization-based cloud computing.Bio: Dr. Jacob Gorm Hansen received his Ph.D. from the University of Copenhagen, Denmark. With Asger Jensen, he was the first to publicly demonstrate live migration of a VM running a full operating system and application stack in 2002, and was an active contributor during the early days of the Xen project led by Cambridge University. He also worked at the University of Washington, designing and building a high-performance and secure display system for "Tahoma", a VM-based browser OS. The Tahoma display system since evolved into "Blink", a 3D-accelerated secure display system for VMs. At VMware since 2007, Dr. Hansen has been working on advanced development projects, primarily the "Lithium" system for scalable and robust VM storage in the cloud. Dr. Hansen lives near Aarhus, Denmark with his wife and infant son. In 2009, he received the ACM EuroSys Roger Needham Award for his thesis and "an exceptional, innovative contribution to knowledge in the systems area." |
| 16.12.2010 | Uwe Erdmann | Fujitsu | Cloud Computing - Trends und Anwendungen in der Industrie (Folien) |
| 06.01.2010 | Keine Veranstaltung | ||
| 13.01.2011 | Dr. Michael Missbach | Hewlett-Packard SAP Competence Center |
SAP as a Service with cloud infrastructures (Folien)Virtualization for SAP and Accounting in Virtualized Environments based on SAPS-MeterSAP as a Service and cloud based services are currently among the most discussed topics. The session shows the current trends in SAP Infrastructures; discussing the promises and challenges of Clouds as well as the myths and reality of Virtualization and demonstrate how the HP Matrix concept in conjunction with the innovative HP SAPS-meter web service can be deployed for SAP cloud Services; helping customers to become green and reduce TCO without jeopardizing their SLA's. |
| 20.01.2011 | Enno Folkerts | SAP AG |
Benchmarking and Performance Prediction in the Cloud (Folien)Traditional performance benchmarks are designed with respect to widely accepted principles such as reproducibility, relevancy, accessibility, scalability and simplicity. A benchmark is defined by a system landscape, software specifications and the rules, how to run the software. It measures the quality of the underlying hardware and software by performance numbers such as throughput or interaction rate. Vendors use the benchmark results to appraise their products. Customers can use the results as a decision help, which vendor to choose and for sizing considerations.We will argue, that the traditional benchmarking principles need to be adjusted in order to make sense in a cloud scenario. Besides scalability a cloud benchmark would necessarily need some elasticity extension. As customers rather buy a service and not hardware or software anymore, the accessibility principle may also have to be redefined. Last but not least, we will see how techniques based on pattern recognition can help building a realistic workload for benchmarks in the cloud. |
| 27.01.2011 | Thijs Metsch | Platform Computing |
Delivering Cloud Services! - Why, How and (some) Use CasesThis talk will try to give an overview of Cloud computing and the corresponding user experience. It shows why NOW is the right time for Cloud solutions. Focus will be on federation and Cloud Interfaces, and why Standards are needed for Interoperability, Innovation, Integration and Portability! The talk will cover the usual Grid, HPC and Cloud uses cases and demonstrate how these might extend into the Cloud. |
| 03.02.2011 | Ingo Schwarzer, CTO | DB Systel GmbH |
Cloud Hype ? Trend ? Fake ? Einordnung in die Unternehmensstrategie (Folien) |
| 10.02.2011 | Achim Reeh | Capgemini sd&m AG |
CapGeminis TechnoVision und Cloud Computing (Folien)Der Vortrag stellt im Rahmen von Capgeminis TechnoVision aktuelle IT-Trends in der Wirtschaft vor und beleuchtet den Einfluss, den Cloud Computing darauf hat. Die Herausforderungen an die Durchführung von Cloud-Projekten werden anhand zweier konkreter Projekte aus dem Software-as-a-Service- und Infrastructure-as-a-Service-Bereich dargestellt. |
| 17.02.2011 | Cataldo Mega | ECM Architecture & Solutions IBM Deutschland Development & Research GmbH |
Software as a Service Modelle in Kontext von Enterprise Content Management - Am Beispiel von e-Archivierung in der Cloud (Folien) |
Die Bewertung erfolgt anhand der allgemein anerkannten Regeln zur Erstellung wissenschaftlicher Publikationen. Dazu gehören u.a. Rechtschreibung und Grammatik, das korrekte Zitieren von Quellenmaterial, Bildnachweise, eine plausible Dokumentstruktur sowie korrektes Schlussfolgern.
Die Ausarbeitungen sind in deutscher Sprache zu erstellen. Ausnahmen sind nur für Nicht-Muttersprachler möglich. Alle Ausarbeitungen müssen anhand der IEEE Dokumentvorlage für Word bzw. LaTex (bare_jrnl.tex) erstellt werden.
Die Abgabe erfolgt als PDF-Datei per eMail an peter.troeger@hpi.uni-potsdam.de. Bei Bedarf werden wir sie um die Word / Latex - Originaldateien bitten.
Deadline ist der 31.3.2011, 23:59 MEZ.