| 1. |
- Eklund, Johan, 1966-, et al.
(author)
-
Delay Penalty during SCTP Handover
- 2011
-
Conference paper (peer-reviewed)abstract
- The rapidly growing interest in untethered Internet connections such as WLAN and 3G/4G mobile connections,calls for intelligent session management, not least in terms of handovers. As part of an effort to develop a SCTP-based session management framework, we are studying ways of improving the SCTP handover delay forreal-time traffic by optimizing the startup delay on the handover-target path. We have developed a theoretical model that predicts the transfer times of SCTP messages during the startup on a new path. This paper validates our model. It shows that the model can be used to predict message transfer times in variable bitrate flows. The paper further employs our model to study the startup delay penalty during handover for the spectrum of network conditions considered relevant for real-time traffic over mobile connections.
|
|
| 2. |
- Eklund, Johan, 1966-, et al.
(author)
-
Impact of slow start on SCTP handover performance
- 2011
-
In: Proc Int Conf Comput Commun Networks ICCCN. - : IEEE conference proceedings. - 9781457706387
-
Conference paper (peer-reviewed)abstract
- The rapidly growing interest in untethered Internet connections, especially in terms of WLAN and 3G/4G mobile connections, calls for intelligent session management: a mobile device should be able to provide a reasonable end-user experience despite location changes, disconnection periods and, not least, handovers. As part of an effort to develop a SCTP-based session management framework that meets these criteria, we are studying ways of improving the SCTP handover delay for real-time traffic; especially the startup delay on the connection between a mobile device and the target access point. To obtain an appreciation of the theoretically feasible gains of optimizing the startup delay on the handover-target path, we have developed a model that predicts the transfer times of SCTP messages during slow start. This paper experimentally validates our model and demonstrates that it could be used to predict the message transfer times in a variable bitrate flow by approximating the variable flow with a constant dito. It also employs our model to obtain an appreciation of the startup delay penalties incured by slow start during handovers in typical mobile, real-time traffic scenarios.
|
|
| 3. |
- Eklund, Johan, 1966-, et al.
(author)
-
Implications of using a Large Initial Congestion Window to Improve mSCTP Handover Delay
- 2012
-
In: MOBILITY 2012 : The Second International Conference on Mobile Services, Resources, and Users. - : IARIA. - 9781612082295 ; , s. 116-121
-
Conference paper (peer-reviewed)abstract
- The currently rather heterogeneous wireless landscape makes handover between different network technologies, so-called vertical handover, a key to a continued success for wireless Internet access. Recently, an extension to the Stream Control Transmission Protocol (SCTP) – the Dynamic Address Reconfiguration (DAR) extension – was standardized by IETF. This extension enables the use of SCTP for vertical handover. Still, the way vertical handover works in SCTP with DAR makes it less suitable for real-time traffic. Particularly, it takes a significant amount of time for the traffic to ramp up to full speed on the handover target path. In this paper, we study the implications of an increased initial congestion window for real-time traffic on the handover target path when competing traffic is present. The results clearly show that an increased initial congestion window could significantly reduce the transfer delay for real-time traffic, provided the fair share of the available capacity on the handover target path is sufficiently higher than the send rate required by the real-time flow. Additionally, we notice that this performance gain comes without penalizing the competing traffic.
|
|
| 4. |
- Eklund, Johan, 1966-, et al.
(author)
-
Improving mSCTP-based Vertical Handovers by Increasing the Initial Congestion Window
- 2011
-
Conference paper (peer-reviewed)abstract
- Wireless networks play an increasingly important role in our daily lives. However, the wireless landscape comprises serveralnetwork technologies, and which technology to use is often context dependent and varies over time: In urban areas, WiFi viaa WiFi hotspot might be the preferred access technology, while in rural areas 3G, HSPA or some other 3GPP technologymight be the only viable alternative. To this end, a large number of vertical handover solutions, ranging from link-level toapplication-level solutions, has been proposed in the past several years.We believe that the Stream Control Transmission Protocol and its extension for Dynamic Address Reconfiguration (mSCTP)make up an attractive vertical handover solution, especially in cases where there are little or no economic incentives to modify orupgrade the existing network infrastructure. Currently, we are implementing an mSCTP-based mobility management frameworkfor Android smartphones and tablets, and as part of that work we are considering ways of mitigating the effects of a handoveron ongoing transport sessions.A mobile terminal with several network interfaces may traverse different types of networks (WLAN, UMTS, HSPA).A seamless handover between these networks may be achieved by the transport protocol, Stream Control TransmissionProtocol(SCTP), with its multihoming facility (mSCTP), in conjunction with a mobility manager monitoring the availablenetwork interfaces. SCTP, a reliable transport protocol, closely related to TCP, encompasses slow-start in the beginning of asession. This slow-start phase is also activated to probe for network capacity on the target path after a handover, which couldresult in unncessarily reduced throughput in case of spare capacity in the network. The throughput degradation in could becrucial for an ongoing real time session, like video conferencing. In this study we have conducted experiments on the benefitof increasing the initial congestion window to reduce the negative impact of the slow start phase. We have by experimentsquantified the performance gain from an increasd initial window, and have seen that there a reduction in maximum transfertime for a message of about 50 % in scenarios with real-time video, compared to the default initial congestion window. Theimprovement is most prominent in sessions with high transfer delays. One easy, and thus very attractive, way to improveon the handover performance is to increase the initial congestion window on the handover target path. Although a seeminglycontroversal proposition, one should bear in mind that researchers at Google and other cloud computing companies are currentlystrongly promoting a drastic increase of TCP’s initial congestion window within IETF.
|
|
| 5. |
- Eklund, Johan, 1966-, et al.
(author)
-
On the Use of an Increased Initial Congestion Window to Improve mSCTP Handover Performance
- 2012
-
In: WAINA 2012. - : IEEE Press. - 9780769546520 ; , s. 1101-1106
-
Conference paper (peer-reviewed)abstract
- With the wireless landscape being ratherheterogeneous, handover between different networktechnologies, so-called vertical handover, becomes keyto a continued success for wireless Internet access.Recently, an extension to the Stream Control Trans-mission Protocol (SCTP) – the Dynamic Address Re-configuration (DAR) extension – was standardized bythe IETF. This extension enables the use of SCTPfor vertical handover. Still, the way vertical handoverworks in SCTP with DAR makes it less suitable for real-time traffic. Particularly, it takes a significant amountof time for the traffic to ramp up to full speed on thehandover target path. In this paper, we study the ex-tent to which an increased initial congestion window onthe handover target path decreases the transfer delayspikes in real-time video traffic experienced during avertical handover. The impact on both standard andhigh-definition video traffic is considered. The resultsof our study suggest that an increased initial congestionwindow does indeed significantly decrease the spikes inthe video traffic. However, the results also indicate thatit does not resolve the problem altogether.
|
|
| 6. |
- Eklund, Johan, 1966-, et al.
(author)
-
Theoretical Analysis of an Ideal Startup Scheme in Multihomed SCTP
- 2010
-
In: Networked Services and Applications - Engineering, Control and Management. - Berlin, Heidelberg : Springer. - 9783642139703 ; , s. 155-166
-
Conference paper (peer-reviewed)abstract
- SCTP congestion control includes the slow-start mechanism to probe the network for available bandwidth. In case of a path switch in a multihomed association, this mechanism may cause a sudden drop in throughput and increased message delays. By estimating the available bandwidth on the alternate path it is possible to utilize a more efficient startup scheme. In this paper, we analytically compare and quantify the degrading impact of slow start in relation to an ideal startup scheme. We consider three different scenarios where a path switch could occur. Further, we identify relevant traffic for these scenarios. Our results point out that the most prominent performance gain is seen for applications generating high traffic loads, like video conferencing. For this traffic, we have seen reductions in transfer time of more than 75% by an ideal startup scheme. Moreover, the results show an increasing impact of an improved startup mechanism with increasing RTTs
|
|
| 7. |
- Söderman, Pehr, et al.
(author)
-
Handover in the Wild : The feasibility of vertical handover in commodity smartphones
- 2013
-
In: Communications (ICC), 2013 IEEE International Conference on. - : IEEE conference proceedings. - 9781467331227 ; , s. 6401-6406
-
Conference paper (peer-reviewed)abstract
- Today commodity mobile devices are equipped withmultiple wireless access technologies. To enable continuous connectivityit is vital that these terminals provide for vertical handover between different technologies. Particularly, they should provide a vertical handover that complies with the timeliness requirements of soft real-time applications. Considering aspects such as cost- and ease-of-deployment, application neutrality, and, not least, the emergence of transport protocols that support multi-homing such as mobile SCTP and multi-path TCP, we think it would be beneficial to handle vertical handover in thetransport layer of the mobile terminal. This paper demonstrates through several real-world experiments, the feasibility of using a lightweight vertical handover scheme in smart mobile terminals for live video streaming. The vertical handover criteria is basedon the received signal strength. Our experiments suggest that thescheme indeed provides for seamless vertical handover at walking speed – our target scenario. However, the experiments also suggest that the scheme gives significant reductions in handovertime, as compared to mobile SCTP without improvements, at higher speeds.
|
|
| 8. |
- Grinnemo, Karl-Johan, 1968-, et al.
(author)
-
On the Use of CMT-SCTP to Improve the Startup Behavior of PSTN Signaling Traffic
- 2013
-
Conference paper (other academic/artistic)abstract
- There is currently work going on at IETF to standardize concurrent multipath transfer, i.e., simultaneous transfer of data over several network paths, for SCTP. This paper studies whether or not SCTP extended with concurrent multipath transfer (CMT-SCTP) could provide a faster startup behavior than standard SCTP. The paper complements previous work on CMT-SCTP, and extends it to PSTN signaling traffic. The paper suggests that CMT-SCTP could give a faster startup behavior over symmetrical paths, i.e., paths with similar bandwidths and round-trip times, but that its behavior is sensitive to differences in round-trip time between the paths. The paper also indicates that signaling traffic properties such as message size and burstiness contribute marginally to the startup behavior of CMT-SCTP.
|
|
| 9. |
- Grinnemo, Karl-Johan, 1968-, et al.
(author)
-
On the Use of MPTCP to Reduce Latency for Cloud Applications
- 2014
-
Conference paper (other academic/artistic)abstract
- As mobile cloud usage has become more and more prevalent, with more people depending on the cloud for both their work and their leisure, cloud-to-end-user latency have risen to become a key issue. One oft-forgotten way to cost effectively improve the quality of experience is to simultaneously make use of the several network interfaces, e.g., WiFi and 3G/4G interfaces, available on most of todays mobile devices including smartphones and tablets; that is, to employ multihoming. This paper provides an initial evaluation of the latency characteristics of the multipath extensions to TCP, Multipath TCP (MPTCP), that are currently being standardized by IETF. In particular, the paper considers the possible reductions in latency that could be obtained by using MPTCP and multiple network paths between a cloud service and a mobile end user. Traffic from three cloud applications are studied, Netflix, Google Maps, and Google Docs, representing typical high-, mid-, and low- intensity cloud-to-end-user traffic. The result suggests that significant latency reductions are indeed possible, however, primarily for high- intensity video traffic such as Netflix.
|
|
| 10. |
|
|
