2008
DOI: 10.1504/ijcnds.2008.020258
|View full text |Cite
|
Sign up to set email alerts
|

Low jitter guaranteed-rate communications for cluster computing systems

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

1
4
0

Year Published

2009
2009
2017
2017

Publication Types

Select...
4
1
1

Relationship

0
6

Authors

Journals

citations
Cited by 8 publications
(5 citation statements)
references
References 27 publications
1
4
0
Order By: Relevance
“…The low-jitter GR scheduling algorithm removes much of the cell delay variability associated with the dynamic scheduling algorithms used in existing IP routers, and minimizes the amount of buffering required in the IP routers. Extensive simulations indicate that essentially-zero delay jitter is achievable, as confirmed by theoretical results [42], [43], [45] and other experimental results [46], [47]. The extensive simulations indicate that each IP router typically buffers 2 cells (128 bytes) of video data per output port for the aggregated IPTV flow being multicast, significantly less buffering than current IP routers require.…”
Section: Discussionsupporting
confidence: 63%
See 3 more Smart Citations
“…The low-jitter GR scheduling algorithm removes much of the cell delay variability associated with the dynamic scheduling algorithms used in existing IP routers, and minimizes the amount of buffering required in the IP routers. Extensive simulations indicate that essentially-zero delay jitter is achievable, as confirmed by theoretical results [42], [43], [45] and other experimental results [46], [47]. The extensive simulations indicate that each IP router typically buffers 2 cells (128 bytes) of video data per output port for the aggregated IPTV flow being multicast, significantly less buffering than current IP routers require.…”
Section: Discussionsupporting
confidence: 63%
“…Many GR scheduling algorithms have been proposed in the literature [32]- [47], and a brief summary is provided in Section III. To date all previous proposed GR schedulers [32]- [41] cannot achieve 100% throughput efficiency while maintaining unity 'switch speedup', and they cannot provide sufficiently small bounds on (a) the number of cells queued in each IQ switch, (b) the queuing delays within an IQ switch, or (c) the delay jitter within an IQ switch given unity speedup.…”
Section: Introductionmentioning
confidence: 99%
See 2 more Smart Citations
“…The same approach has been used to create novel packet schedulers for specific wireless links. This is the case of Cohen and Grebla (2014) and Grebla et al (2015) for the LTE technology, (Cicconetti et al, 2014;Ke et al, 2014) for WiMAX technology, and (Szymanski and Gilbert, 2008) for cluster systems. In these works, the synergy between the selected packet scheduler and the medium technology is even higher, considering characteristics bounded and tailored for the specific problem under exam.…”
Section: Related Workmentioning
confidence: 99%