Core Services Goals for 2005-2006
Goals for the current planning year
Goals for previous years
Overview
Substantial areas of infrastructure work this year will include
completion of the networked storage installation for DLPS resources,
and work toward broadening the infrastructure for general web services;
rollout of the CoSign authentication system for all web services; and
introduction of redundant MySQL service.
Newly-hired programming staff will enable long-standing goals of
replacing the DLPS statistics system and completion of the
authorization management system to be completed, as well as new
automation for the Michigan Digitization Project and other projects to
be handled as well.
Goals are organized below into three areas:
New and improved services, Infrastructure, and
Work flow/process/organizational.
- New and improved services
- Michigan Digitization Project
(MDP) access system. Support the development of the MDP access
system and develop and improve automation for MDP workflows.
- Priority: HIGH
- Specifics/timeline:
- Winter: Develop fully-automated download management system.
- Winter/Spring: Build additional infrastructure for MDP
access system.
- Institutional Repository.
Support the Institutional Repository pilot project.
- Shibboleth. Collaborate
with ITCS to begin to offer Shibboleth origin service for UM.
- Priority: MEDIUM
- Specifics/timeline:
- Winter/Spring: Coordinate ITCS work with ProQuest toward a
pilot.
- Summer: Release ("soft") pilot for ProQuest resources,
troubleshoot, operationalize.
- Fall: Announce production support for Shibboleth access to
ProQuest resources, begin planning for broading service to other
resource providers.
- Infrastructure
- Improvements to storage
infrastructure, web server redundancy. Continue migration of DLPS
resources to new model, extend to library web service, and modernize
load-balancing mechanisms.
- Priority: HIGH
- Specifics/timeline:
- Winter: Colocate historically distinct DLPS services
(ETS, HTI, Image) onto networked storage and web server farm to
facilitate data synchronization.
- Spring: Investigate synchronization issues with library
web service and research options for modernizing load balancing.
- Summer/Fall: Migrate library web service to networked
storage and web server farm and implement improvements to load
balancing.
- COSIGN. Transition from
Lib Auth to COSIGN authentication
- Priority: HIGH
- Specifics/timeline:
- Winter/Spring: Colocate historically distinct DLPS
services (ETS, HTI, Image) onto networked storage and web server farm
to facilitate SSL configuration and coordinate with Web Services to
rework
web application logic as necessary to work with COSIGN.
- Spring: Develop
rollout plan in consultation with PARC.
- Summer: Make final
preparations for rollout and cut over.
- Redundant MySQL. Combine
production MySQL database services to a central, dedicated, redundant
pair of servers.
- Priority: HIGH
- Specifics/timeline:
- Winter: Research,
activate, and test redundancy, develop backup
procedures, and migrate
DLPS databases to new servers.
- Spring: Migrate library web databases to new servers.
- Production server
replacements/upgrades.
- Priority: HIGH
- Specifics/timeline:
- Migrate from CAEN AFS to ITCS AFS.
- Winter: Complete pending
2004-2005 server upgrades and migrations of
unsupported RedHat versions.
- After Winter: Initiate 2005-2006
server and storage upgrades.
- Replace DLXS
statistics.
Research off-the-shelf solutions that can be easily adapted to handle
DLXS and MDP statistics and that can be sustainably managed and
released via DLXS.
- Priority: HIGH
- Specifics/timeline:
- Spring: Research available tools and development of
customized modules.
- Summer: Transition from current system, including
migration of data as possible.
- Develop an authentication and
authorization management system.
- Priority: HIGH
- Specifics/timeline:
- Spring/Summer: Design management features.
- Summer/Fall: Development and rollout of new features.
- Centralize LIT backup.
Migrate departmental CS and DSS backup services to a larger-scale,
centralized backup infrastructure, purchasing or re-using equipment as
appropriate, to reduce costs and improve scale.
- Priority: MEDIUM
- Specifics/timeline:
- Winter: Test hardware, determine capacities/costs,
and develop architecture.
- Spring: Purchase software and transition service.
- Formalize system
administration infrastructure.
- Priority: MEDIUM
- Specifics:
- Define and document the
characteristics of the standard environment: what platforms and
operating systems are supported, what software packages are installed,
which of those are centrally administered, etc.
- Explore tools and
brainstorm models for centralized system administration, including
techniques for automating the processes of installing operating systems
and base required software.
- Set up system
administration resources such as servers, software, etc. as necessary.
- Begin transition of
servers to centralized model.
- Revisit logging practives.
Conduct a basic audit of logging practices in LIT systems and work
toward better compliance with Library privacy policy.
- Priority: MEDIUM
- Specifics/timeline:
- Winter: Create
inventory of systems and degree of compliance.
- Spring: Formulate
strategies for coming into compliance.
- Ongoing: Implement
strategies, where possible.
- DHCP/DNS services.
Migrate existing DHCP services to ITCS-provided HindSite service.
- Priority: MEDIUM
- Specifics:
- Rollout to the LIT
Hatcher private network.
- Rollout to the general
library networks.
- Development server
replacements/upgrades. Conduct periodic assessments of development
servers and identify areas of need. With input from users, consider a)
consolidating services, b) repurposing retired production servers (if
still viable), and/or c) purchasing new equipment or upgrades as
appropriate. Work to achieve a balance between a robust environment and
management overhead.
- Priority: LOW
- Specifics:
- Migrate from CAEN AFS to ITCS AFS.
- Security assessment.
Conduct an assessment of security practices, document them, and make
improvements where necessary.
- Work flow, processes, and
organizational work
- Trusted Digital Repository (TDR) compliance and quality
management. The new repository, access
system, and associated work flows and documentation for the MDP
project are the sole focus of this planning year's TDR work. However,
critical quality management needs in digital reformatting work flows
are also to be addressed.
- Priority: MEDIUM
- Specifics/timeline:
- Improve or develop
automation in the digital reformatting unit for the following:
- file name convention checking
- image validation (using JHove)
- MD5 checksum creation/validation
- Evaluate
MDP-related processes with respect to TDR compliance, using the new RLG
auditing checklist. (To be coordinated by new LIT Special Projects
Librarian)
- Position new MDP repository for TDR compliance.
- Stage 1 (after February, to avoid competition with
development of prototype access system): Share results of above
evaluation, and identify critical areas or opportunities in new
repository architecture, work flow, and/or documentation during early
formative periods.
- Stage 2: Build components, set up work flows, or
write documentation in accordance with TDR principles as identified in
Stage 1.
- Regular validation of digital objects. Develop
automation
for regularly validating digital objects using MD5 checksums.
- Persistent IDs and
resolution. Support the development of processes to create persistent
URLs for DLPS- and SPO-hosted material.
- Documentation and
standardization of processes. Organize and fill gaps in
documentation of the current environment and normal operations; use the
process of documenting processes as an opportunity to develop and
implement stanadards for those processes.
- Improve coordination of data
loading. Establish communications as appropriate to better
coordinate demand for data loading.
- LIT work requests and
scheduling. Improve communication of work
requests across LIT, including scheduling.
- LIT on-call service.
Further streamline LIT on-call service.
- Cost model. Continue
development of the model for core service costs based on a simple,
known variable such as page or image count, to better quantify
resources needed in Core Services.
- Technology trends awareness.
Pursue a modest number of options such as trade magazine subscriptions
or technology conferences to stay abreast of development in relevant
technologies.