Exogenous-Loss Awareness in Queue Management: Toward Global Fairness

For a given TCP flow, exogenous losses are those occurring on links other than the flow's bottleneck link. Exogenous losses are typically viewed as introducing undesirable "noise" into TCP's feedback control loop, leading to inefficient network utilization and potentially severe global unfairness. This has prompted much research on mechanisms for hiding such losses from end-points. In this paper, we show through analysis and simulations that low levels of exogenous losses are surprisingly beneficial in that they improve stability and convergence, without sacrificing efficiency. Based on this, we argue that exogenous loss awareness should be taken into account in any AQM design that aims to achieve global fairness. To that end, we propose an exogenous-loss aware Queue Management (XQM) that actively accounts for and leverages exogenous losses. We use an equation based approach to derive the quiescent loss rate for a connection based on the connection's profile and its global fair share. In contrast to other queue management techniques, XQM ensures that a connection sees its quiescent loss rate, not only by complementing already existing exogenous losses, but also by actively hiding exogenous losses, if necessary, to achieve global fairness. We establish the advantages of exogenous-loss awareness using extensive simulations in which, we contrast the performance of XQM to that of a host of traditional exogenous-loss unaware AQM techniques.

Bayesian Packet Loss Detection for TCP

One of TCP's critical tasks is to determine which packets are lost in the network, as a basis for control actions (flow control and packet retransmission). Modern TCP implementations use two mechanisms: timeout, and fast retransmit. Detection via timeout is necessarily a time-consuming operation; fast retransmit, while much quicker, is only effective for a small fraction of packet losses. In this paper we consider the problem of packet loss detection in TCP more generally. We concentrate on the fact that TCP's control actions are necessarily triggered by inference of packet loss, rather than conclusive knowledge. This suggests that one might analyze TCP's packet loss detection in a standard inferencing framework based on probability of detection and probability of false alarm. This paper makes two contributions to that end: First, we study an example of more general packet loss inference, namely optimal Bayesian packet loss detection based on round trip time. We show that for long-lived flows, it is frequently possible to achieve high detection probability and low false alarm probability based on measured round trip time. Second, we construct an analytic performance model that incorporates general packet loss inference into TCP. We show that for realistic detection and false alarm probabilities (as are achievable via our Bayesian detector) and for moderate packet loss rates, the use of more general packet loss inference in TCP can improve throughput by as much as 25%.

Using Speculation to Reduce Server Load and Service Time on the WWW

Speculative service implies that a client's request for a document is serviced by sending, in addition to the document requested, a number of other documents (or pointers thereto) that the server speculates will be requested by the client in the near future. This speculation is based on statistical information that the server maintains for each document it serves. The notion of speculative service is analogous to prefetching, which is used to improve cache performance in distributed/parallel shared memory systems, with the exception that servers (not clients) control when and what to prefetch. Using trace simulations based on the logs of our departmental HTTP server http://cs-www.bu.edu, we show that both server load and service time could be reduced considerably, if speculative service is used. This is above and beyond what is currently achievable using client-side caching [3] and server-side dissemination [2]. We identify a number of parameters that could be used to fine-tune the level of speculation performed by the server.

Load Profiling in Distributed Real-Time Systems

Load balancing is often used to ensure that nodes in a distributed systems are equally loaded. In this paper, we show that for real-time systems, load balancing is not desirable. In particular, we propose a new load-profiling strategy that allows the nodes of a distributed system to be unequally loaded. Using load profiling, the system attempts to distribute the load amongst its nodes so as to maximize the chances of finding a node that would satisfy the computational needs of incoming real-time tasks. To that end, we describe and evaluate a distributed load-profiling protocol for dynamically scheduling time-constrained tasks in a loosely-coupled distributed environment. When a task is submitted to a node, the scheduling software tries to schedule the task locally so as to meet its deadline. If that is not feasible, it tries to locate another node where this could be done with a high probability of success, while attempting to maintain an overall load profile for the system. Nodes in the system inform each other about their state using a combination of multicasting and gossiping. The performance of the proposed protocol is evaluated via simulation, and is contrasted to other dynamic scheduling protocols for real-time distributed systems. Based on our findings, we argue that keeping a diverse availability profile and using passive bidding (through gossiping) are both advantageous to distributed scheduling for real-time systems.

Load Profiling for Efficient Route Selection in Multi-Class Networks

High-speed networks, such as ATM networks, are expected to support diverse Quality of Service (QoS) constraints, including real-time QoS guarantees. Real-time QoS is required by many applications such as those that involve voice and video communication. To support such services, routing algorithms that allow applications to reserve the needed bandwidth over a Virtual Circuit (VC) have been proposed. Commonly, these bandwidth-reservation algorithms assign VCs to routes using the least-loaded concept, and thus result in balancing the load over the set of all candidate routes. In this paper, we show that for such reservation-based protocols|which allow for the exclusive use of a preset fraction of a resource's bandwidth for an extended period of time-load balancing is not desirable as it results in resource fragmentation, which adversely affects the likelihood of accepting new reservations. In particular, we show that load-balancing VC routing algorithms are not appropriate when the main objective of the routing protocol is to increase the probability of finding routes that satisfy incoming VC requests, as opposed to equalizing the bandwidth utilization along the various routes. We present an on-line VC routing scheme that is based on the concept of "load profiling", which allows a distribution of "available" bandwidth across a set of candidate routes to match the characteristics of incoming VC QoS requests. We show the effectiveness of our load-profiling approach when compared to traditional load-balancing and load-packing VC routing schemes.

Evaluation of a Load Profiling Approach to Routing Guaranteed Bandwidth Flows

To support the diverse Quality of Service (QoS) requirements of real-time (e.g. audio/video) applications in integrated services networks, several routing algorithms that allow for the reservation of the needed bandwidth over a Virtual Circuit (VC) established on one of several candidate routes have been proposed. Traditionally, such routing is done using the least-loaded concept, and thus results in balancing the load across the set of candidate routes. In a recent study, we have established the inadequacy of this load balancing practice and proposed the use of load profiling as an alternative. Load profiling techniques allow the distribution of "available" bandwidth across a set of candidate routes to match the characteristics of incoming VC QoS requests. In this paper we thoroughly characterize the performance of VC routing using load profiling and contrast it to routing using load balancing and load packing. We do so both analytically and via extensive simulations of multi-class traffic routing in Virtual Path (VP) based networks. Our findings confirm that for routing guaranteed bandwidth flows in VP networks, load balancing is not desirable as it results in VP bandwidth fragmentation, which adversely affects the likelihood of accepting new VC requests. This fragmentation is more pronounced when the granularity of VC requests is large. Typically, this occurs when a common VC is established to carry the aggregate traffic flow of many high-bandwidth real-time sources. For VP-based networks, our simulation results show that our load-profiling VC routing scheme performs better or as well as the traditional load-balancing VC routing in terms of revenue under both skewed and uniform workloads. Furthermore, load-profiling routing improves routing fairness by proactively increasing the chances of admitting high-bandwidth connections.

Task Assignment in a Distributed System: Improving Performance by Unbalancing Load

We consider the problem of task assignment in a distributed system (such as a distributed Web server) in which task sizes are drawn from a heavy-tailed distribution. Many task assignment algorithms are based on the heuristic that balancing the load at the server hosts will result in optimal performance. We show this conventional wisdom is less true when the task size distribution is heavy-tailed (as is the case for Web file sizes). We introduce a new task assignment policy, called Size Interval Task Assignment with Variable Load (SITA-V). SITA-V purposely operates the server hosts at different loads, and directs smaller tasks to the lighter-loaded hosts. The result is that SITA-V provably decreases the mean task slowdown by significant factors (up to 1000 or more) where the more heavy-tailed the workload, the greater the improvement factor. We evaluate the tradeoff between improvement in slowdown and increase in waiting time in a system using SITA-V, and show conditions under which SITA-V represents a particularly appealing policy. We conclude with a discussion of the use of SITA-V in a distributed Web server, and show that it is attractive because it has a simple implementation which requires no communication from the server hosts back to the task router.

End-to-End Inference of Loss Nature in a Hybrid Wired/Wireless Environment

End-to-End differentiation between wireless and congestion loss can equip TCP control so it operates effectively in a hybrid wired/wireless environment. Our approach integrates two techniques: packet loss pairs (PLP) and Hidden Markov Modeling (HMM). A packet loss pair is formed by two back-to-back packets, where one packet is lost while the second packet is successfully received. The purpose is for the second packet to carry the state of the network path, namely the round trip time (RTT), at the time the other packet is lost. Under realistic conditions, PLP provides strong differentiation between congestion and wireless type of loss based on distinguishable RTT distributions. An HMM is then trained so observed RTTs can be mapped to model states that represent either congestion loss or wireless loss. Extensive simulations confirm the accuracy of our HMM-based technique in classifying the cause of a packet loss. We also show the superiority of our technique over the Vegas predictor, which was recently found to perform best and which exemplifies other existing loss labeling techniques.

Effectiveness of Loss labeling in Improving TCP Performance in Wired/Wireless Networks

The current congestion-oriented design of TCP hinders its ability to perform well in hybrid wireless/wired networks. We propose a new improvement on TCP NewReno (NewReno-FF) using a new loss labeling technique to discriminate wireless from congestion losses. The proposed technique is based on the estimation of average and variance of the round trip time using a filter cal led Flip Flop filter that is augmented with history information. We show the comparative performance of TCP NewReno, NewReno-FF, and TCP Westwood through extensive simulations. We study the fundamental gains and limits using TCP NewReno with varying Loss Labeling accuracy (NewReno-LL) as a benchmark. Lastly our investigation opens up important research directions. First, there is a need for a finer grained classification of losses (even within congestion and wireless losses) for TCP in heterogeneous networks. Second, it is essential to develop an appropriate control strategy for recovery after the correct classification of a packet loss.

Simple Load Balancing for Distributed Hash Tables

Distributed hash tables have recently become a useful building block for a variety of distributed applications. However, current schemes based upon consistent hashing require both considerable implementation complexity and substantial storage overhead to achieve desired load balancing goals. We argue in this paper that these goals can b e achieved more simply and more cost-effectively. First, we suggest the direct application of the "power of two choices" paradigm, whereby an item is stored at the less loaded of two (or more) random alternatives. We then consider how associating a small constant number of hash values with a key can naturally b e extended to support other load balancing methods, including load-stealing or load-shedding schemes, as well as providing natural fault-tolerance mechanisms.

Entropy Loss is Maximal for Uniform Inputs

A secure sketch (defined by Dodis et al.) is an algorithm that on an input w produces an output s such that w can be reconstructed given its noisy version w' and s. Security is defined in terms of two parameters m and m˜ : if w comes from a distribution of entropy m, then a secure sketch guarantees that the distribution of w conditioned on s has entropy m˜ , where λ = m−m˜ is called the entropy loss. In this note we show that the entropy loss of any secure sketch (or, more generally, any randomized algorithm) on any distribution is no more than it is on the uniform distribution.

TCP over CDMA2000 Networks: A Cross-Layer Measurement Study

Modern cellular channels in 3G networks incorporate sophisticated power control and dynamic rate adaptation which can have a significant impact on adaptive transport layer protocols, such as TCP. Though there exists studies that have evaluated the performance of TCP over such networks, they are based solely on observations at the transport layer and hence have no visibility into the impact of lower layer dynamics, which are a key characteristic of these networks. In this work, we present a detailed characterization of TCP behavior based on cross-layer measurement of transport, as well as RF and MAC layer parameters. In particular, through a series of active TCP/UDP experiments and measurement of the relevant variables at all three layers, we characterize both, the wireless scheduler in a commercial CDMA2000 network and its impact on TCP dynamics. Somewhat surprisingly, our findings indicate that the wireless scheduler is mostly insensitive to channel quality and sector load over short timescales and is mainly affected by the transport layer data rate. Furthermore, we empirically demonstrate the impact of the wireless scheduler on various TCP parameters such as the round trip time, throughput and packet loss rate.

Load Balancing a Cluster of Web Servers using Distributed Packet Rewriting

In this paper, we propose and evaluate an implementation of a prototype scalable web server. The prototype consists of a load-balanced cluster of hosts that collectively accept and service TCP connections. The host IP addresses are advertised using the Round Robin DNS technique, allowing any host to receive requests from any client. Once a client attempts to establish a TCP connection with one of the hosts, a decision is made as to whether or not the connection should be redirected to a different host---namely, the host with the lowest number of established connections. We use the low-overhead Distributed Packet Rewriting (DPR) technique to redirect TCP connections. In our prototype, each host keeps information about connections in hash tables and linked lists. Every time a packet arrives, it is examined to see if it has to be redirected or not. Load information is maintained using periodic broadcasts amongst the cluster hosts.

Unicast-based Characterization of Network Loss Topologies

Current Internet transport protocols make end-to-end measurements and maintain per-connection state to regulate the use of shared network resources. When a number of such connections share a common endpoint, that endpoint has the opportunity to correlate these end-to-end measurements to better diagnose and control the use of shared resources. A valuable characterization of such shared resources is the "loss topology". From the perspective of a server with concurrent connections to multiple clients, the loss topology is a logical tree rooted at the server in which edges represent lossy paths between a pair of internal network nodes. We develop an end-to-end unicast packet probing technique and an associated analytical framework to: (1) infer loss topologies, (2) identify loss rates of links in an existing loss topology, and (3) augment a topology to incorporate the arrival of a new connection. Correct, efficient inference of loss topology information enables new techniques for aggregate congestion control, QoS admission control, connection scheduling and mirror site selection. Our extensive simulation results demonstrate that our approach is robust in terms of its accuracy and convergence over a wide range of network conditions.

TCP over CDMA2000 Networks: A Cross-Layer Measurement Study

Modern cellular channels in 3G networks incorporate sophisticated power control and dynamic rate adaptation which can have significant impact on adaptive transport layer protocols, such as TCP. Though there exists studies that have evaluated the performance of TCP over such networks, they are based solely on observations at the transport layer and hence have no visibility into the impact of lower layer dynamics, which are a key characteristic of these networks. In this work, we present a detailed characterization of TCP behavior based on cross-layer measurement of transport layer, as well as RF and MAC layer parameters. In particular, through a series of active TCP/UDP experiments and measurement of the relevant variables at all three layers, we characterize both, the wireless scheduler and the radio link protocol in a commercial CDMA2000 network and assess their impact on TCP dynamics. Somewhat surprisingly, our findings indicate that the wireless scheduler is mostly insensitive to channel quality and sector load over short timescales and is mainly affected by the transport layer data rate. Furthermore, with the help of a robust correlation measure, Normalized Mutual Information, we were able to quantify the impact of the wireless scheduler and the radio link protocol on various TCP parameters such as the round trip time, throughput and packet loss rate.