The abstracts of the technical talks are sorted by topic/headline.
Robert M. Raymond
Recent introductions of new tape drives with extremely high capacity cartridges and announcements of future technologies capable of even higher capacities is creating a need to use tape differently. This need has created ideas for revolutionary features and functions that lay the foundation for better tape management, performance and data integrity. For example, a new tape format was introduced providing 100s of linked partitions enabling tape to be used more like disk. A new standard for an enterprise class data integrity validation has been developed which is able to detect and report data corruption before the data even gets written to tape. A new features enabling a drive to do a string search through the written data on the tape to find and report the locations of all the matches has been developed. This presentation will describe some of these new ideas for revolutionary features and how they might be used in various applications.
Jos van Wezel
Widespread available computing power and network bandwidth resulted in an explosion of data growth and information in our society, economy and science. The amount of data generated by instruments, experiments, measurements, scanning, or direct production in all disciplines of natural sciences as well as humanities is increasing fast. Scientific communities demand a proper storage, management, handling and analysis of their scientific data with powerful tools, services, and systems. Most of the scientific data is an extremely valuable good, as re-producing it is not possible; hence the need for long-term archival is evident and mandatory. At the Steinbuch Centre for Computing of the Karlsruhe Institute of Technology the storage and archive systems of the Large Scale Data Facility allow a multi-disciplinary use which will scale to multiple PetaBytes. In the LSDF, TSM is used for long time storage of millions of files. The presentation describes the goal of the LSDF its architecture and discusses future challenges for long time archival.
Alexander Dunaevskiy
A concept of database backup for TSM is not only the pure or scheduled TSM command “BACKUP DB”. Typically such concept contains a set of different TSM schedules like “Start/Stop of RECLAIMS and MIGRATIONS”, “BACKUP STG”, “DELETE VOLHISTORY” additionally “DEVICE CLASSES” and “SERVER GROUPS” and finally schedules for “BACKUP DB” if need to be to different “DEVICE CLASSES”.
The complexity of such scheme causes by adjusts of parameter of a single schedule and possible interactions with other schedules. The whole system is sensitive to the changes.
The change of the internal database from proprietary database to DB2 leads to many problems for existing concepts on large TSM systems. We will discuss the influence of the internal database change on the concept of database backup for TSM and experiences at Leibniz Supercomputing Centre.
This paper will be useful for a TSM administrator, who would like to build a well defined database protection for a farm of TSM server instances.
Greg van Hise
This session is a follow-on to "Data Protection in Virtual Server Environments - Introducing TSM for Virtual Environments" and will provide a deep dive into Data Protection for VMware. Best practices for sizing, configuration and administration will be discussed. Specific discussion topics include: node naming recommendations, scheduling, tape, deduplication and security considerations. Deployment options for the B/A client on a vStorage backup server and the DP for VMware Recovery agent will be reviewed.
Jim Smith, Greg Van Hise
Virtual machines have different characteristics than physical machines that must be considered when planning backup and recovery. Given these differences, there are a number of methodologies that can be employed to meet recovery time and recovery point objectives. This session will discuss different types of virtualization (server, desktop, application) and explore options available for virtual machine backup/recovery with Tivoli Storage Manager. We will discuss general approaches for virtual machine protection and introduce TSM for Virtual Environments - Data Protection for VMware (concepts, terminology, and components).
Matt Anglin
This presentation will examine the various deduplication choices in Tivoli Storage Manager. The session will compare client-side deduplication with server-side deduplication and external deduplication, such as ProtectTIER. Planning for deduplication in terms of database and storage pool space will be covered. The speaker will discuss various data types and how best to deduplicate them, and examine server options to help control deduplication in your environment.
Ralf Müller, Peter Micke
A lot of TSM Administrators are facing a list of failed and missed client schedules every day, which is a time consuming problem. Some of the TSM events fail because of known problems on the client side. Re-valuation of the results might help solving the problem. Backup Eagle provides information about events after a revaluation and documents them in a revision fair way. Besides that a documentation about every change in the TSM server configuration is done. All of these features help to lower the cost of daily administration tasks and let the admin focus on other TSM or non-TSM problems.
In the presentation we talk about a customer installation
Bruno Friess
In today’s infrastructure TSM has become a very important solution for data of critical applications. TSM has become a 24x7 application, but is still treated as a “backup program”. Simple operating system clusters are build to keep TSM alive. But is this really enough? How much does a cluster know of TSM? Does it make sense to run TSM as a cluster-aware application?
Jim Smith
The Tivoli Storage Manager Backup-Archive client includes a wide variety of options and techniques for backing up your data. This session will address specific scenarios in which these methods should be considered for protecting large file systems. This will include discussion of journal-based backup for Windows and AIX file/web servers, snapshot-assisted backup for NetApp filers and GPFS, and use of images for backup and recovery.
Matt Anglin
This session will explore the TSM V6 database from deployment and performance recommendations through best practices and troubleshooting. The deployment discussion will cover performance and configuration recommendations for new and upgraded servers. Server upgrade capabilities and considerations will also be covered. This session will provide details about the server database, lessons learned since 2009 when the new database was introduced, and other hints and tips for a successful experience.
Mike Bobo
Update on the latest in innovations in the Support, Services, and Education areas from IBM. Review of latest additions and strategies for in self-healing initiatives, simplified self help, and just in time training and education as well as status of initiative to provide seamless storage hardware/software support from IBM. Guidance on finding latest best practices documents, maintenance and update strategy, and pointers on how to gain the most from maintenance and subscription spending. Specific guidance on improvements coming in 2012 and beyond that will help customers gain more value from the resources within the IBM Tivoli Support, Services, and Education teams.
Discussion of the roadmap for Tivoli Support, Services, and Education for the next 3 years. Review of the enhancements to electronic support, base and premium support, and education offerings. Overview of changes to support model, service delivery, and education and training improvements. Open forum for discussion of Support and Services issues and actions taken by IBM to address those issues as brought forward by the user community over the last 2 years. Ample time will be allocated for questions and open discussion.
Matt Anglin
This presentation will explore various clustering techniques for the Tivoli Storage Manager server, primarily on Windows, AIX, and Linux. It will look at how to configure clustering products such as Tivoli Systems Automation, PowerHA (formerly HACMP), and MSCS. Sample cluster setups will be explored on Windows and AIX in particular, focusing on how to set up (and how not to set up) DB2. The presentation will also cover other forms of high availability, including DB2 HADR and Export/Import as ways of protecting a server from disaster. Characteristics of each technique will be explored, including how long it takes to fail over, and how various cluster solutions work with disasters of varying scope (system outage, storage sub-system outage, site outage).
Thomas Eifert
Users ask for restore. In this talk the activities at RWTH Aachen are depicted that are made to ensure the ability to restore data in a fast, reliable and cost-effective way.
Dave Cannon
This session provides a detailed preview of selected future enhancements in the next TSM release and beyond. The speaker will discuss potential improvements in replication, global data management, administration, protection of VMware data, scalability, security and compliance.
Josef Weingand
This session will discuss how to architecture a tailor-made solution including several and different Storage Tiers, like Disk, Tape, VTL and DataDeDuplication.
I will discuss the pros and cons of different storage tiers and how to utilize the advantages of it for different data classes in combination with TSM. I give you examples when and for which data to use tape, TSM Disk Pool, VTL, or Data DeDuplication, but I show you also how to combine all of them in a Smarter and tailor-made Backup Solution.
Josef Weingand
I will cover technical news on IBM Tape, VTL and Data DeDuplication and the implementation and usage considerations for TSM, such as:
Josef Weingand
Do you think tape is dead? I will show you why tape, even it is nearly 60 years old, is not dead and will not die in future. I discuss the current technology and the current development, but also the advantage of tape storage over other storage tiers. New function, like the IBM Tape File System (LTFS), will lead to a renaissance of tape.
Del Hoobler, Andreas Uhl
This presentation provides a general overview of FlashCopy Manager for Windows and FlashCopy Manager for UNIX and Linux platforms and its value proposition for protecting business-critical applications in your enterprise. It takes a deeper dive into FlashCopy Manager for Windows and how it utilizes Microsoft's Volume Shadow Copy Service (VSS) architecture to protect Microsoft SQL Server and Microsoft Exchange Server.
Del Hoobler
This presentation provides a general overview of IBM Tivoli Storage Manager for Mail (Data Protection for Domino and Data Protection for Exchange) and how to use it to protect IBM Lotus Domino servers or Microsoft Exchange servers in your enterprise. It then takes a deeper dive into Data Protection for Exchange, especially as it relates to the new paradigms of Microsoft Exchange Server 2010 and Exchange Individual Mailbox Restore ("IMR").
Tom O’Brien
Come to this session to get an update on the Tivoli Storage Manager (TSM) roadmap. TSM Product Management will discuss IBM's current product direction for the short (and long) terms. This discussion will include a high-level view of the different functions that will be delivered in 2011 such as plans for additional disaster recovery scenarios, support for virtual environments, TSM's UI and more. Time will be allowed for Q&A. A follow-up session will discuss the technical implementation details for some of these future functions.
Oren Wolf
Please join us to learn about recent updates to the TSM portfolio. This session will cover the various products included in the TSM portfolio and enhancements that have been delivered over the last 12-18 months. TSM Product Manager will discuss IBM's perspective on the most common challenges that TSM customers are facing and how recent enhancements contribute to a unified data protection strategy to address those challenges. Some of the topics that will be discussed are: cross-product integration points, deduplication (source and target), automatic client updates, the latest TDP release (TSM for Virtual Environments), support for hardware snapshots through Flashcopy manager, recent updates to the licensing model and more. Time will be allowed for Q&A.
Kathy Mitton
This session will explore TSM server operations, daily maintenance, and best practices to optimize your TSM server. The speaker will also discuss the administrative and reporting capabilities for the server along with examples and rationale for managing and scheduling server maintenance tasks.
Urs Moser, IBM Global Services SSO TE/Winterthur
TSM migration (V5.5 to 6.2) in large environments
Including Library Sharing (7x 3485)
Or
TSM migration (TSM on z/OS) / into new TSM Server 6.2 on LINUX RedHat (seven TSM instance under one HW on each side)
Virtual Library sharing within two side and NDMP backup into the virtual library environment with ACLS between the two sides
Daniele Francesco Kruse
The TSM server network at CERN - with its 17 TSM servers in production, 30 drives, ~1300 client nodes and ~4 PB of data - often needs an overwhelming amount of effort to be properly managed by the few TSM administrators. Thus, the need for a central monitoring system able to cope with the increasing number of servers, client nodes and volumes. We will discuss our approach to this issue, focusing on TSMMS, a TSM Monitoring System developed inhouse, able to give an effective view of the needs and status of the network and of the individual servers, as well as statistics and usage reports. Avoiding repetitive error-prone manual checks, TSM admins are able to manage the whole TSM system just by looking at the periodic reports, and taking appropriate action. TSMMS scales seamlessly with the enlargement of the network, thus saving the cost of additional administrative personnel.
Remco Post
This presentation will be about my experiences with TSM version 6 fail-over on AIX, using home-grown scripts, thus avoiding HACMP.
My presentation will cover:
Norbert Pott
Hana Shparber
Lothar Wollschläger
Our space managed GPFS file system contains over 40 Million files.
Therefore the candidate search for migration with the scoutd and the backup of this file system takes too long. Using the GPFS policy engine is a solution for space management. The GPFS Policy engine will be described and how to use it for space management with TSM/HSM. Also the performance gain will be outlined.
Because also the backup for this file system takes too long the use of mmbackup came under consideration. Mmbackup also use the policy engine and TSM for backing up a GPFS file system. The problems with mmbackup will be described and also the performance gain with mmbackup.
The last topic will be our future considerations for disaster recovery with the use of mmbackup and space management for all our GPFS home file systems.
