Transcript Telemicroscopy Session Model

User and Network Interplay in Internet Telemicroscopy
Prasad Calyam (Presenter)
Nathan Howes, Mark Haffner, Abdul Kalash
Ohio Supercomputer Center, The Ohio State University
IMMERSCOM, October 11th 2007
Topics of Discussion
 Telemicroscopy Overview
 Motivation
 Use-cases
 Solutions
 Telemicroscopy Session Model
 User and Network Interplay
 Testbed for Experiments to Characterize Model Parameters
 Performance Analysis
 OSC’s Remote Instrumentation Collaboration Environment (RICE)
 Features
 Demo Video
 Conclusion
2
Telemicroscopy Overview
 Academia and Industry use computer-controlled scientific instruments
 Electron Microscopes, NMR, Raman Spectrometers, Nuclear Accelerator
 For research and training purposes
 Cancer Cure, Material Science, Nanotechnology
 Instruments are expensive ($450K - $ 4Million) and need dedicated staff to
maintain
+) Remote instrumentation benefits




Access to users who cannot afford to buy instruments
Return on Investment (ROI) for instrument labs
Avoids duplication of instrument investments for funding agencies (NSF, OBOR)
Useful when physical presence of humans around sample is undesirable
-) Remote instrumentation drawbacks
 Improper operation can cause physical damages that are expensive to repair
Telemicroscopy is remote instrumentation of electron microscopes
3
Telemicroscopy Use-cases
 Tele-observation versus Tele-operation
4
Telemicroscopy Solutions
 Hardware-based: KVM over IP (KVMoIP)
 Encoder-Decoder pair for frame-differencing based video image transfers
 Pros: High quality video and optimal response times
 Cons: Expensive, Special hardware and high-end bandwidth
requirements
 Software-based: VNC – remote desktop software
 Raw or copy-rectangle or JPEG/MPEG encoded video image transfers
 Pros: Inexpensive, Easily deployable
 Cons: Improper PC hardware or network congestion can degrade video
quality and optimal control response times
5
Related Work
 Telemicroscopy over Internet2
 Gemini Observatory
 NanoManipulator
 Telescience Project – National Center for Microscopy
and Imaging Research, UC San Diego
 Ultrahigh Voltage Electron Microscope Research
Center – Osaka University
 Common Instrument Middleware Architecture (CIMA)
– Indiana University
 Tele-presence Microscopy – Argonne National Lab’s
Advanced Analytical Electron Microscope facility
+) Novel applications for
controlling instruments
+) All said “it works” over
XYZ network paths and
listed challenges they
overcame
-) None have quantified
performance in terms of
network effects
-) None have considered
user Quality of
Experience (QoE)
Study Motivation: Understanding User and Network interplay can help us improve
reliability and efficiency of Telemicroscopy and thus deliver optimum user QoE
6
Telemicroscopy Session Model
(a) Session Model Parameters
(b) Closed-loop Control System Representation
→ user-activity (key strokes and mouse clicks) during a
session involving n microscope functions
→ average video image transfer rate at the microscope end
→ network connection quality
→ input-output scaling factor; unique to a microscope function
→ seed image transfer rate; for quick screen refresh
→ average video image transfer rate at the user end
→ system-state control parameter dependent on user
behavior; causes ± feedback in the control system
7
Telemicroscopy Session Model
(a) Session Model Parameters
(b) Closed-loop Control System Representation
(c) Transfer Function
(d) End-user QoE relation in a Telemicroscopy session
 Demand – Effort the user had to expend to perform n actions
 Supply – Perceivable video image quality during the n actions
8
Telemicroscopy System States
(Effects of H parameter)
(a) State Transitions
(b) System Supply-Demand Performance
9
Case Study: OSC Collaboration with OSU CAMM
 OSU Center for Accelerated Maturation of Materials (CAMM)
has acquired high-end Electron Microscopes
 Used for materials modeling studies at sub-angstrom level
 OSC providing systems and networking support for
Telemicroscopy
 OSCnet supporting end-to-end bandwidth requirements
 Image processing of samples (automation with MATLAB) for
Analytics service
 Telemicroscopy Demonstrations
 Supercomputing, Tampa, FL (Nov 2006)
 Internet2 Fall Member Meeting, Chicago, IL (Dec 2006)
 Stark State University/Timken, Canton, OH (Mar 2007)
10
Telemicroscopy Testbed
 Experiments to characterize session model parameters
 Test cases with different network connections – CAMM requirements
 (a) 1 Gbps LAN (Direct connection to Users in neighboring room)
 (b) Isolated LAN (Users in the same building )
 (c) Public LAN (Users in different buildings on campus)
 (d) WAN (Users on the Internet)
 Performance analysis goals
 Bandwidth, latency and packet loss levels for optimum user QoE
 Traffic characterization for studying inter-play between user control
(TCP traffic) and microscope response (UDP traffic)
11
WAN Testbed
(a) Setup
(b) WAN Path Performance
12
Performance Measurements Collected
End-user QoE Measurements (Subjective Metrics)



Mean Opinion Scores (MOS) of “Novice” and “Expert” Users
Time for completion of “basic” and “advanced” Tele-microscopy
tasks by Novice and Expert Users
Network Measurements (Objective Metrics)


Collected using Ethereal/TCPdump and OSC ActiveMon
 Metrics: Data rate, Protocols Summary
13
Network Connection Quality (ψnet) and User QoE (qmos)
 qmos notably decreases with decrease in network connection quality
 User QoE is highly sensitive to network health fluctuations
 Novice more liberal than Expert
 Time taken to complete a task increases with decrease in network connection
quality
NOTE: qmos of 5 corresponds to “at the microscope” QoE
14
Network Connection Quality (ψnet) and User Control (bin)
 Mouse and Keyboard traffic is TCP traffic
 Higher TCP throughput on poor network connections
 Increased user effort with keyboard and mouse on poor connections
 “Congestion begets more congestion”
1 Gbps LAN – Expert
Task-1
60 B/s
Task-2
Task-3
Public 100 Mbps LAN – Expert
Task-1
900 B/s
Task-2
Task-3
100 Mbps WAN – Expert
Task-1
Task-2
Task-3
1400 B/s
60 s
User expends minimum
effort with keyboard and
mouse to complete use-case
100 s
User expends notably more
effort with keyboard and mouse
to complete use-case
15
140 s
User expends a “lot” of effort
with keyboard and mouse to
complete use-case
Network Connection Quality (ψnet) and Image Transfer Rate (Δbout)
 “At the microscope” QoE requires ~30 Mbps between user and microscope ends
 Other WAN tests at SC06 (Tampa) and Internet2 FallMM (Chicago) to
microscopes at CAMM (Columbus)
 Usable on ~(10-25) Mbps WAN connections
 Usable if one-way network delays within ~50ms; as much as ~20% UDP packet loss
tolerable if adequate bandwidth provisioned
16
OSC’s Remote Instrumentation Collabration
Environment (RICE)
 Leverages our user and network interplay studies for “reliable” and “efficient”
Telemicroscopy sessions and thus delivers optimum user QoE
 Customizable software on custom server-side hardware for Telemicroscopy
 Best of VNC and KVMoIP worlds
 RICE Features
 Network-aware video encoding
 Optimizes frame rates based on available network bandwidth
 Manual video-quality adjustment slider
 Network-status and user-action blocking
 Warns user of network congestion that leads to unstable session state
 Blocks user-actions during extreme congestion scenarios and prevents
breakdown
 Collaboration tools
 VoIP, Chat, Annotation, Command-abstraction
 Multi-user support
 Control-lock passing, collaborators presence, colored-text chat conference
 Workflow and Image management
 Simultaneously connects to multiple PCs, transfers images and
transparently switches between them
17
RICE Demo Video
18
RICE use-cases for online learning
 Remote students can view instructor (also remote!) controlling different
types of scientific instruments
 Efficiently – with the appropriate video frames to match last mile network
capabilities
 Reliably – without worrying about damaging the instrument
 Multi-party VoIP and Chat collaboration
 Image Annotation
 Instructor can pass control to students - train them to operate the instrument
during the class
 Students can conduct lab sessions at their assigned slots on the
instruments
 Students image files can be organized and hosted at a central server
 Analytics can be supported using a web-service to analyze the image data sets
19
Future Work
 Shared instrumentation uses OSC’s state-wide resources
 Networking, Storage, HPC, Analytics
Cyberinfrastructure for Shared Instrumentation
20
Shared Instrumentation @ OSC
 Plans underway to support shared instrumentation for  Ohio State University: CAMM Electron Microscopes, Chemistry
Department Spectrometers and Diffractometers, Astronomy
Department Telescopes
 Miami University: Electron Microscopes, EPR Spectrometers
 Ohio University: Nuclear Accelerator
21
Thank you for your attention!
☺
Any Questions?
22