Saqib Raza

Network level footprints of facebook applications

With over half a billion users, Online Social Networks (OSNs) are the major new applications on the Internet. Little information is available on the network impact of OSNs, although there is every expectation that the volume and diversity of traffic due to OSNs is set to explode. In this paper, we examine the specific role played by a key component of OSNs: the extremely popular and widespread set of third-party applications on some of the most popular OSNs. With over 81,000 third-party applications on Facebook alone, their impact is hard to predict and even harder to study. We have developed and launched a number of Facebook applications, of which are among the most popular applications on Facebook in active use by several million users monthly. Through our applications, we are able to gather, analyze, correlate, and report their workload characteristics and performance from the perspective of the application servers. Coupled with PlanetLab experiments, where active probes are sent through Facebook to access a set of diverse applications, we are able to study how Facebook forwarding/processing of requests/responses impacts the overall delay performance perceived by end-users. These insights help provide guidelines for OSNs and application developers. We have also made the data studied here publicly available to the research community. This is the first and only known study of popular third-party applications on OSNs at this depth.

Graceful network state migrations

A significant fraction of network events (such as topology or route changes) and the resulting performance degradation stem from premeditated network management and operational tasks. This paper introduces a general class of Graceful Network State Migration (GNSM) problems, where the goal is to discover the optimal sequence of operations that progressively transition the network from its initial to a desired final state while minimizing the overall performance disruption. We investigate two specific GNSM problems: 1) Link Weight Reassignment Scheduling (LWRS) studies the optimal ordering of link weight updates to migrate from an existing to a new link weight assignment; and 2) Link Maintenance Scheduling (LMS) looks at how to schedule link deactivations and subsequent reactivations for maintenance purposes. LWRS and LMS are both combinatorial optimization problems. We use dynamic programming to find the optimal solutions when the problem size is small, and leverage ant colony optimization to get near-optimal solutions for large problem sizes. Our simulation study reveals that judiciously ordering network operations can achieve significant performance gains. Our GNSM solution framework is generic and applies to similar problems with different operational contexts, underlying network protocols or mechanisms, and performance metrics.

Network Coding Based Cooperative Peer-to-Peer Repair in Wireless Ad-Hoc Networks

Cooperative Peer-to-Peer Repair (CPR) has been proposed to recover from packet losses incurred during 3G broadcast. CPR leverages the increasing presence of multi-homed mobile devices having both 3G cellular and IEEE 802.11 wireless interfaces. Mobile devices can, therefore, draw upon IEEE 802.11 peering links to cooperatively achieve out-of-band repair of 3G broadcasting losses. This paper considers the problem of employing Network Coding (NC) to exploit the broadcast nature of the wireless medium towards enhancing the efficiency of CPR. We show that the minimum latency scheduling problem for NC based CPR (NC-CPR) is NP-Hard. We present heuristics for NC-CPR that assume a priori topology and packet loss information. Insights gained from our heuristics are leveraged to propose NC-DCPR, a fully distributed protocol for NC-CPR. We conduct extensive simulation experiments under realistic network conditions. Our results show that employing network coding significantly improves the efficiency of CPR.

Cooperative Peer-to-Peer Repair for Wireless Multimedia Broadcast

This paper explores how to leverage IEEE802.11-based cooperative peer-to-peer repair (CPR) to enhance the reliability of wireless multimedia broadcasting. We first formulate the CPR problem and present an algorithm that assumes global state information to optimally schedule CPR transmissions. Based on insights gained from the optimal algorithm, we propose a fully distributed CPR (DCPR) protocol. Simulation results demonstrate that the DCPR protocol can effectively enhance the reliability of wireless broadcast services with a repair latency comparable to that of optimal scheduling.

MeasuRouting: a framework for routing assisted traffic monitoring

Monitoring transit traffic at one or more points in a network is of interest to network operators for reasons of traffic accounting, debugging or troubleshooting, forensics, and traffic engineering. Previous research in the area has focused on deriving a placement of monitors across the network toward the end of maximizing the monitoring utility of the network operator for a given traffic routing. However, both traffic characteristics and measurement objectives can dynamically change over time, rendering a previously optimal placement of monitors suboptimal. It is not feasible to dynamically redeploy/reconfigure measurement infrastructure to cater to such evolving measurement requirements. We address this problem by strategically routing traffic subpopulations over fixed monitors. We refer to this approach as MeasuRouting. The main challenge for MeasuRouting is to work within the constraints of existing intradomain traffic engineering operations that are geared for efficiently utilizing bandwidth resources, or meeting quality-of-service (QoS) constraints, or both. A fundamental feature of intradomain routing, which makes MeasuRouting feasible, is that intradomain routing is often specified for aggregate flows. MeasuRouting can therefore differentially route components of an aggregate flow while ensuring that the aggregate placement is compliant to original traffic engineering objectives. In this paper, we present a theoretical framework for MeasuRouting. Furthermore, as proofs of concept, we present synthetic and practical monitoring applications to showcase the utility enhancement achieved with MeasuRouting.

Dynamic measurement-aware routing in practice

Traffic monitoring is a critical network operation for the purpose of traffic accounting, debugging or troubleshooting, forensics, and traffic engineering. Existing techniques for traffic monitoring, however, tend to be suboptimal due to poor choice of monitor location or constantly evolving monitoring objectives and traffic characteristics. One way to counteract these limitations is to use routing as a degree of freedom to enhance monitoring efficacy, which we refer to as measurement-aware routing. Traffic sub-populations can be routed (rerouted) on the fly to optimally leverage existing monitoring infrastructures. Implementing dynamic measurement-aware routing (DMR) in practice is riddled with challenges. Three major challenges are how to dynamically assess the importance of traffic flows; how to aggregate flows (and hence take a common action for them) in order to conserve routing table entries; and how to achieve traffic routing/rerouting in a manner that is least disruptive to normal network performance while maximizing the measurement utility. This article takes a closer look at these challenges and discusses how they manifest for different types of networks. Through an OpenFlow prototype, we show how DMR can be applied in enterprise networks. Using global iceberg detection and capture as a driving application, we demonstrate how our solutions successfully route suspected iceberg flows to a DPI box for further processing, while preserving balanced load distribution in the overall network.

Online routing of bandwidth guaranteed paths with local restoration using optimized aggregate usage information

We investigate the problem of distributed online routing of bandwidth guaranteed paths with local restoration. A unified model is proposed that captures the bandwidth sharing characteristic of backup paths that provision local restoration, corresponding to different fault models. We apply the model to describe bandwidth sharing on backup paths for varying degrees of network state information. The extent of backup bandwidth sharing depends on the amount of network state information made available through routing protocols. A key design criterion for traffic engineering schemes is to maximize the sharing between backup paths, while minimizing this protocol overhead. M.S. Kodialam and T.V. Lakshman (see Proc. Infocom, p.376-85, 2001) demonstrated that propagating a constant amount of aggregated information per link leads to cost effective bandwidth sharing. We propose oAIS, a new aggregate information scenario, in which we judiciously select the propagated information, such that the protocol overhead is identical to that of Kodialam and Lakshman. Simulations show that oAIS outperforms other information scenarios with comparable protocol overheads.

RED-BL: Energy solution for loading data centers

Cloud infrastructure providers and data center operators spend a major portion of their operations budget on the electric bills. We present RED-BL (Relocate Energy Demand to Better Locations), a framework for determining an optimal mapping of workload to an existing set of data centers while considering the cost of workload relocation. Within each workload mapping interval, RED-BL solution exploits the geo diversity in electricity price markets. The temporal diversity in those markets is simultaneously exploited by considering a planning window comprising several mapping intervals. Using workload traces from live Internet applications and electricity prices from the US markets, RED-BL can reduce the electric bill by as much as 81% from the case when the workload is equally distributed. Compared to a single data center deployment, an average reduction of 27% in electric bill can be achieved when RED-BL uses 10 or more data centers, a common case for most operators. When compared to existing workload relocation solutions, RED-BL achieves a further reduction of 13.63%, on average. While modest, this reduction can save millions of dollars for the operators. The cost of this saving is an inexpensive computation at the start of each planning window.

MeasuRouting: A Framework for Routing Assisted Traffic Monitoring

Monitoring transit traffic at one or more points in a network is of interest to network operators for reasons of traffic accounting, debugging or troubleshooting, forensics, and traffic engineering. Previous research in the area has focused on deriving a placement of monitors across the network towards the end of maximizing the monitoring utility of the network operator for a given traffic routing. However, both traffic characteristics and measurement objectives can dynamically change over time, rendering a previously optimal placement of monitors suboptimal. It is not feasible to dynamically redeploy/reconfigure measurement infrastructure to cater to such evolving measurement requirements. We address this problem by strategically routing traffic sub-populations over fixed monitors. We refer to this approach as MeasuRouting. The main challenge for MeasuRouting is to work within the constraints of existing intra-domain traffic engineering operations that are geared for efficiently utilizing bandwidth resources, or meeting Quality of Service (QoS) constraints, or both. A fundamental feature of intra-domain routing, that makes MeasuRouting feasible, is that intra-domain routing is often specified for aggregate flows. MeasuRouting, can therefore, differentially route components of an aggregate flow while ensuring that the aggregate placement is compliant to original traffic engineering objectives. In this paper we present a theoretical framework for MeasuRouting. Furthermore, as proofs-of-concept, we present synthetic and practical monitoring applications to showcase the utility enhancement achieved with MeasuRouting.

Graceful Network Operations

A significant fraction of network events (such as topology or route changes) and the resulting performance degradation stem from premeditated network management and operational tasks. This paper introduces a general class of Graceful Network Operation (GNO) problems, where the goal is to discover the optimal sequence of operations that progressively transition the network from its initial to a desired final state while minimizing the overall performance disruption. We investigate two specific GNO problems: (a) Link Weight Reassignment Scheduling (LWRS) studies the optimal ordering of link weight updates to migrate from an existing to a new link weight assign- ment, and (b) Link Maintenance Scheduling (LMS) looks at how to schedule link deactivations and subsequent reactivations for maintenance purposes. LWRS and LMS are both combinatorial optimization problems. We use dynamic programming to find the optimal solutions when the problem size is small, and leverage Ants Colony Optimization to get near-optimal solutions for large problem sizes. Our simulation study reveals that judiciously ordering network operations can achieve significant performance gains. Our GNO solution framework is generic and applies to similar problems with different operational contexts, underlying network protocols or mechanisms, and performance metrics. I. INTRODUCTION The Internet has been an enabling technology for mission- critical applications and services such as Voice over IP, VPNs, e-commerce applications, and multimedia streaming. Such applications rely upon consistent Quality of Service (QoS) provisioning by Internet Service Providers (ISPs), with five-nines availability (99.999% uptime) becoming the norm rather than the exception. The end-to-end perceived QoS can potentially be affected due to the dynamic nature of the networks. For instance, network topology may change due to transient router/link outages or long-term network engineering. Furthermore, protocol configuration parameters may be altered to migrate from one setting to another. Ideally, QoS guarantees should persist across such dynamic conditions. Some of these dynamic changes are inadvertent e.g., ones due to faulty interfaces, router crashes, and accidental fiber cuts. However, other changes ensue from deliberate and pre- meditated actions of network operators (e.g., routine main- tenance). A failure characterization study of the Sprint IP backbone (1) observed that planned maintenance activities account for more than 20% of transient failures. Other studies (2) have also observed the prevalence of such planned main- tenance activities. Premeditated network tasks also include network upgrade activities such as adding new routers to the network and overhauling link capacity. Another example of a premeditated network task is migrating an existing OSPF (3) or IS-IS (4) 1 link weight assignment to a new assignment that has been optimized based upon the most up-to-date traffic matrix estimates. In the case of premeditated tasks, network operators have the prerogative to decide the sequence of atomic operations that comprise such a task. This paper introduces a general class of problems referred to as Graceful Network Operation (GNO) problems, which typically involve migrating a network from its initial state to a final state by executing a series of atomic operations. Each of these operations may cause some performance disruption that is a function of the networks state. The GNO problem is to discover the sequence of operations that progressively transition the network to the final state while minimizing the overall disruption. This paper looks at two specific GNO problems, as described below.