Dell Precision R5400 Remote Access Device: Task-Based User Guide  - Page 21

Latency Considerations

Page 21 highlights

PCoIP Technology User Guide (Vol I) 9 Latency Considerations With any network, there are latency effects to be addressed. The following are latency effect considerations for PCoIP Technology deployments: • Latency effects are subjective and affect response (e.g. mouse movement) • Latency due to length of physical medium (i.e. speed of light of copper/fiber) and switch hops • Additional latency due to OS (e.g. 40-50 ms for Windows) The table below outlines latency effects examples based on critical user evaluation. User tolerance levels for latency vary widely. Table 2: Latency Network Latency Approx. Distance1 0-30 ms Campus/Metro/ Inter-city (0-1500 km) 40-60 ms Inter-city/Intracountry (1500-2500 km) 60-100 ms Intra-country/ Intercontinent (2500-5000 km) > 100 ms Inter-continent/ Overseas (> 5000 km) Example Observations2 Perception free to average user Minimal latency perceived, e.g. 'heavier' than usual mouse pointer/windows, but very usable Sluggish mouse and windows Some audio/ video dropouts Slow mouse and windows Audio/video dropouts Notes: 1. High bandwidth, low error network 2. Examples given are based on subjective critical analysis Note: As with bandwidth considerations, these latency observations are subjective and biased towards a perception free experience. Administrators must study use case(s) typical for their deployment and adjust user expectations accordingly. TER0806003 Issue 1 Desired Network Attributes The desired network attributes to minimize latency effects are: • High bandwidth • Low error rate • Minimized data path/network hops Networks with less desirable network attributes will result in decreased performance, i.e. decreased mouse and display responsiveness. USB Latency Performance USB performance can vary over latency for the various transfer types: • Isochronous - may notice delay or loss of data (e.g. video data lost on a webcam) • Interrupt - may delay device response (e.g. slow keystrokes from keyboard) • Bulk - may notice slower data transfer (e.g. slower USB FLASH drive) Minimizing Latency Effects The following are suggestions for extreme long distances, non-enterprise networks or less than desirable networks to minimize latency effects: • Use graphically less demanding computer applications and set user expectations accordingly • For bandwidth constricted environments, configure Device Target Bandwidth and Device Bandwidth Limit parameters to limit bandwidth usage to minimum required (see Section 6 Bandwidth Considerations) • Configure Image settings to minimum image quality required (see Section 7 Imaging Considerations). © 2008 Teradici Corporation Page 21 of 25

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25

PCoIP Technology User Guide (Vol I)
TER0806003 Issue 1
© 2008 Teradici Corporation
Page 21 of 25
9 Latency Considerations
With any network, there are latency effects to be
addressed. The following are latency effect
considerations for PCoIP Technology
deployments:
Latency effects are subjective and affect
response (e.g. mouse movement)
Latency due to length of physical medium (i.e.
speed of light of copper/fiber) and switch hops
Additional latency due to OS (e.g. 40-50 ms
for Windows)
The table below outlines latency effects
examples based on critical user evaluation. User
tolerance levels for latency vary widely.
Table 2: Latency
Network
Latency
Approx. Distance
1
Example
Observations
2
0-30 ms
Campus/Metro/
Inter-city
(0-1500 km)
Perception free to
average user
40-60 ms
Inter-city/Intra-
country
(1500-2500 km)
Minimal latency
perceived, e.g.
‘heavier’ than usual
mouse
pointer/windows,
but very usable
60-100 ms
Intra-country/ Inter-
continent
(2500-5000 km)
Sluggish mouse
and windows
Some audio/ video
dropouts
> 100 ms
Inter-continent/
Overseas
(> 5000 km)
Slow mouse and
windows
Audio/video
dropouts
Notes
:
1. High bandwidth, low error network
2. Examples given are based on subjective
critical analysis
Note: As with bandwidth considerations, these
latency observations are subjective and biased
towards a perception free experience.
Administrators must study use case(s) typical for
their deployment and adjust user expectations
accordingly.
Desired Network Attributes
The desired network attributes to minimize
latency effects are:
High bandwidth
Low error rate
Minimized data path/network hops
Networks with less desirable network attributes
will result in decreased performance, i.e.
decreased mouse and display responsiveness.
USB Latency Performance
USB performance can vary over latency for the
various transfer types:
Isochronous – may notice delay or loss of data
(e.g. video data lost on a webcam)
Interrupt – may delay device response (e.g.
slow keystrokes from keyboard)
Bulk – may notice slower data transfer (e.g.
slower USB FLASH drive)
Minimizing Latency Effects
The following are suggestions for extreme long
distances, non-enterprise networks or less than
desirable networks to minimize latency effects:
Use graphically less demanding computer
applications and set user expectations
accordingly
For bandwidth constricted environments,
configure Device Target Bandwidth and
Device Bandwidth Limit parameters to limit
bandwidth usage to minimum required (see
Section 6 Bandwidth Considerations)
Configure Image settings to minimum image
quality required (see Section 7 Imaging
Considerations).