Multimedia Storage Management

1
Multimedia Storage Management

• Sunil PrabhakarAssistant ProfessorDepartment of
  Computer SciencesPurdue University

2
Researchers

• W. Aref (CS)
• A. Elmagarmid (CS)
• A. Ghafoor (ECE)
• R. L. Kashyap (ECE)
• M. Khan (ECE)
• S. Prabhakar (CS)

3
On-Going Research Projects for Multimedia Storage
Management

• Efficient image storage
• Disk placement schemes
• Tertiary storage management
• I/O scheduling and reliability for robotic
  libraries
• Multimedia document management
• Document placement techniques
• I/O management
• Goal improved performance

4
Image Placement for Multiresolution Browsing
Images decomposed into multiple resolutions
using wavelets
5
Image Placement on Disks

• Content-based retrieval using browsing graphs or
  quantitative measures
• Multidimensional data declustering for similarity
  queries
• Designed and evaluated (through simulation and
  implementation) several efficient disk placement
  schemes

6
Tertiary Storage Management

• Scheduling and reliability for tertiary storage
  are critical issues. These issues arise due to
  high latency and rapid tape wear-out
• Most scheduling work has focussed on single
  medium scheduling
• Work on reliability has been limited to mirroring
  or RAIT

7
Research Results for Tertiary Storage Management

• Scheduling
• Developed scheduling algorithms
• Optimal for single drive libraries
• Near-optimal for multiple drive libraries
• Significant improvement over FCFS
• Tape reliability
• Tape Group Parity Protection
• Lower storage overhead than mirroring
• Better performance than RAIT

8
Challenges for Multimedia Document Placement

• Mixed media storage (audio, video, text, images)
• Mixed media data placement techniques
• Statistical techniques using access patterns
• Document striping across multiple disks
• Document clustering and allocation techniques for
  parallel disk arrays and clusters of networked
  servers

9
Multimedia Browsing Graph
Abomasum
Omasum
Reticulum
Protozoa
Esophagus
Rumen
Stomach
Digestive System
Small Intestine
Salivary Glands
Large Intestine
10
I/O Management Issues

• I/O channels between various levels of the
  storage hierarchy are limited resources
• To ensure QoS at run-time intelligent I/O channel
  allocation schemes are needed.
• Document profiles can be utilized for dynamic I/O
  channel allocation

11
Document Bandwidth Profile
12
Related Work

• Image storage
• Chabot (UC Berkeley)
• QBIC (IBM)
• Multimedia storage
• Fellini (Lucent)
• Mitra (USC)
• IBM (Phillip Yu et al)
• NASD
• Streaming RAID (Tobagi et al)

13
Planned Multi-Level Storage Architecture
Environments
Clustered Server Architecture
Parallel Disk Array Architecture
RAID
Multimedia Document Profiles Traces
Multimedia Document Profiles Traces
14
Experiments for Multi-Level Storage Architecture
Environments

• Objective Comparison between the two
  environments in terms of QoS guarantees
• Real-time scheduling
• Multimedia data placement schemes at different
  levels of the storage hierarchy
• I/O channel management
• Reliability
• Scalability