180 Park Ave - Building 103
Florham Park, NJ
SMOG: A Cloud Platform for Seamless Wide Area Migration of Networked Games
Jeffrey Pang, Seungjoon Lee, Jacobus Van , Virajith Jalaparti, Matthew Caesar
The 11th Annual Workshop on Network and Systems Support for Games,
2012.
[PDF]
[BIB]
IEEE Copyright
This version of the work is reprinted here with permission of IEEE for your personal use. Not for redistribution. The definitive version was published in 2012 , 2012-11-22
{Highly interactive network applications such as online games are rapidly growing in popularity but remain challenging for game providers to support, due to their inherent need for low latency. While cloud computing has proven a useful infrastructure for other applications, existing cloud computing facilities are insufficient for games, due to the unpredictability of their workload, their demands on latency and scale, and the need to support game-specific requirements (e.g., players may wish to play with certain other players already in the game). In this work, we explore whether dynamic optimization of latency and scaling of games can be achieved by supplementing cloud computing infrastructure with seamless wide area virtual machine migration using network based route control. We propose SMOG, a framework that dynamically migrates game servers to their optimal location, and uses orchestrated route control to optimize the network path to the server to minimize observable effects of live server migration. Through deployment of a prototype implementation on a Tier-1 ISP’s backbone and a user study, we found SMOG can decrease average end-user latency by up to 60% while performing migration in a manner transparent to game players. While this paper’s focus is online games, SMOG is general enough to be used for a variety of latency-sensitive interactive applications such as video conferencing and interactive video streaming.}
Path Inference in Data Center Networks
Aman Shaikh, Vijay Gopalakrishnan, Seungjoon Lee, Kyriaki Levanti, Hyong Kim, Emmanuil Mavrogiorgis
CNSM (Conference on Network and Service Management),
2012.
[PDF]
[BIB]
IFIP Copyright
The definitive version was published in 2012. , 2012-10-22, http://www.cnsm-conf.org/2012/index.html
{Data center networks have started to play increasingly important roles
in today's Internet. Popular web-based services and critical
enterprise applications are hosted in large data centers. More recent
advances like cloud computing and cellular-based data usage have only
increased the importance of data centers. With the increasing
importance, however, also comes increasing complexity. Supporting the
wide-array of applications and traffic types while meeting all
their performance and security requirements results in complex network
designs. The result of this complexity is that managing these networks
has never been more difficult.
In this paper, we focus on providing one of the key building blocks of
network management: the ability to determine how traffic flows in the
network. This information is fundamental to many different network
management tasks including troubleshooting, capacity
planning, and what-if analysis. Towards that end, we present Chartis, a
system which performs per-packet path inference in a data
center. Chartis takes as input device configurations and the
network's physical topology and outputs the path of a packet in the
network. Specifically, Chartis performs per-hop path inference based on a
simplified model of layer-3 routing, layer-2 switching, and the most
commonly used routing and forwarding mechanisms.
To show Chartis diverse applicability, we perform path inference
within a campus network and on multiple data centers serving the 3G
network of a major cellular service provider. Using routing information
collected from these networks, we validate the correctness of the inferred
paths. Our results show that Chartis can quickly and accurately
determine paths traversed by packets even in complex data center
networks, making it a valuable addition to a network operator's
toolbox.}
MPLS-TE in Networks With Varying Capacity Constraints
Kyriaki Levanti, Vijay Gopalakrishnan, Hyong S. Kim, Seungjoon Lee, Aman Shaikh
Kyriaki Levanti's PhD Dissertation,
2012.
[LINK]
[BIB]
Kyriaki Levanti Copyright
Kyriaki Levanti's PhD Dissertation
{}
3G Meets the Internet -- Understanding Performance Issues due to Hierarchical Routing in 3G Networks
Seungjoon Lee, Zihui Ge, University of Texas Wei Dong
ITC 2011,
2011.
[PDF]
[BIB]
ITC Copyright
The definitive version was published in ITC 2011. , 2011-09-06, http://i-teletraffic.org/fileadmin/ITC23_files/Copyright_form_ITC2011.pdf
{The volume of Internet traffic over 3G wireless networks is sharply rising.
In contrast to many Internet services utilizing replicated resources
(e.g., content distribution networks), the current 3G standard
architecture employs
hierarchical routing, where all user data traffic goes through a small
number
of aggregation points using logical tunnels.
In this paper, we investigate the potential system inefficiency and
performance issues due to
the interplay of the two systems.
We first identify a number of aspects affecting system inefficiency and
service performance and then quantify the impact by analyzing trace data
obtained from a large-scale 3G network and a CDN provider.
We find that extra packet headers used for hierarchical routing
result in significant byte overhead (around 6\%).
We also find that the detour due to hierarchical routing can cause a
packet to travel
extra distance by up to 1627km on the average case, which based on
our data analysis, corresponds to around 45.4\% increase in round-trip
latency.
Furthermore, we identify a pathological case due to DNS caching when a
mobile
device switches between wireless technologies.
In our measurement study on the Internet, we find that this issue can cause
up to an order of magnitude throughput degradation (0.9Mbps vs. 10.8Mbps).
We also study how to achieve performance improvement by deploying
system resources more strategically.}
Optimal Content Placement for a Large-Scale VoD System
David Applegate, Aaron Archer, Vijay Gopalakrishnan, Seungjoon Lee, Kadangode Ramakrishnan
Proceedings of ACM CoNext 2010,
Submitted to ACM CoNext 2010,
2010.
[PDF]
[BIB]
ACM Copyright
(c) ACM, 2010. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in ACM CoNext 2010 , 2010-12-01
IPTV service providers offering Video-on-Demand (VoD) typically have many servers at each metropolitan office to store all the videos in the library. With the rapid increase in the VoD library size, it will soon become infeasible to replicate the entire library at each office. We present an approach for intelligent content placement that scales to large VoD library sizes (e.g., 100Ks of videos). We formulate the problem as a mixed integer program (MIP) that takes into account constraints
such as disk space, link bandwidth, and the skew in content popularity. To overcome the challenges of scale, we employ a Lagrangian relaxation-based decomposition technique that can find a near-optimal solution (e.g., within 1-2%) with orders of magnitude speedup, relative to solving even the LP relaxation via standard software. We also present simple strategies to address practical issues such as popularity estimation, content updates, short-term popularity fluctuation, and frequency
of placement updates. Using traces from an operational system, we show that our approach significantly outperforms simpler placement strategies. For instance, our MIP-based solution can serve all requests using only half the link bandwidth used by LRU cache replacement policy. We also investigate the trade-off between disk space and network bandwidth.
Mitigating Low-Rate Denial-Of-Service Attacks In Packet-Switched Networks,
Tue May 14 17:26:18 EDT 2013
A method includes determining, at a network routing device, an average packet drop rate for a plurality of aggregations of packet flows. The method also determines a threshold packet drop rate based on the average packet drop rate, a current packet drop rate for a select aggregation of the plurality of aggregations, and whether at least one packet flow of the select aggregation is potentially subject to a denial-of-service attack based on a comparison of the current packet drop rate to the threshold packet drop rate.
Method For Encoding Frame Data,
Tue May 10 16:05:05 EDT 2011
In applications where data is transmitted in frames of symbols and the transmission medium is such that the probability of correct reception of symbols is, on the average, not uniform for different symbols in a frame, transmission of test frames enables creation of information about the different probabilities of correct reception, and that information is employed by the transmitter to control the manner in which symbols are transmitted so as to ameliorate the effects of the different probabilities of correct reception.
Broadcast Latency Optimization In Multihop Wireless Networks,
Tue Mar 08 16:04:35 EST 2011
A system and methodology that can optimize practical performance by reducing broadcast latency and collisions in a multihop wireless network is provided. The system includes an optimization component that determines a schedule for transmitting and/or receiving messages at nodes in the network during one-to-all broadcasting, all-to-all broadcasting, and/or all-to-one collection. The schedule enables each node in wireless network to participate in broadcasting as soon as possible, such that, when a node transmits a message to its parent to deliver the message to a source node, the children of the node also receive the message and initiate broadcasting the message in their subtrees in parallel. Further, during data collection, the schedule for messages to be transmitted to a sink node is determined based in part on a priority associated with the message. Additionally, the generated schedule does not affect a set of theoretical bounds associated with a broadcast algorithm employed by the network.
