Buffer Management

Destination-based forwarding in traditional IP routers has not been able to take full advantage of multiple paths that frequently exist in Internet Service Provider Networks. As a result, the networks may not operate efficiently, especially when the traffic patterns are dynamic. This paper describes a multipath adaptive traffic engineering mechanism, called MATE, which is targeted for switched networks such as MultiProtocol Label Switching (MPLS) networks. The main goal of MATE is to avoid network congestion by adaptively balancing the load among multiple paths based on measurement and analysis of path congestion. MATE adopts a minimalist approach in that intermediate nodes are not required to perform traffic engineering or measurements besides normal packet forwarding. Moreover, MATE does not impose any particular scheduling, buffer management, or a priori traffic characterization on the nodes. This paper presents an analytical model, derives a class of MATE algorithms, and proves their convergence. Several practical design techniques to implement MATE are described. Simulation results are provided to illustrate the efficacy of MATE under various network scenarios.

In most computer systems, software overhead dominates the cost of messaging, reducing delivered performance, especially for short messages. Efficient software messaging layers are needed to deliver the hardware performance to the application level and to support tightly-coupled workstation clusters. Illinois Fast Messages (FM) 1.0 is a high speed messaging layer that delivers low latency and high bandwidth for short messages. For 128-byte packets, FM achieves bandwidths of 16.2 MB/s and one-way latencies 32 s on Myrinet-connected SPARCstations (user-level to user-level). For shorter packets, we have measured one-way latencies of 25 s, and for larger packets, bandwidth as high as to 19.6 MB/s-delivered bandwidth greater than OC-3. FM is also superior to the Myrinet API messaging layer, not just in terms of latency and usable bandwidth, but also in terms of the message half-power point (n 1 2

Easy to deploy, robust, and highly resilient to failures, epidemic algorithms are a potentially effective mechanism for propagating information in large peer-to-peer systems deployed on Internet or ad hoc networks. It is possible to adjust the parameters of epidemic algorithm to achieve high reliability despite process crashes and disconnections, packet losses, and a dynamic network topology. Although researchers have used epidemic algorithms in applications such as failure detection, data aggregation, resource discovery and monitoring, and database replication, their general applicability to practical, Internet-wide systems remains open to question. We describe four key problems: membership maintenance, network awareness, buffer management, and message filtering, and suggest some preliminary approaches to address them.

Delay Tolerant Networks are wireless networks where disconnections may occur frequently due to propagation phenomena, node mobility, and power outages. Propagation delays may also be long due to the operational environment (e.g. deep space, underwater). In order to achieve data delivery in such challenging networking environments, researchers have proposed the use of store-carry-and-forward protocols: there, a node may store a message in its buffer and carry it along for long periods of time, until an appropriate forwarding opportunity arises. Additionally, multiple message replicas are often propagated to increase delivery probability. This combination of long-term storage and replication imposes a high storage overhead on untethered nodes (e.g. handhelds). Thus, efficient buffer management policies are necessary to decide which messages should be discarded, when node buffers are operated close to their capacity.

The future broadband ISDN needs a high degree of flexibility in order to cope with a great variety of services with widely differing bandwidth and quality of service requirements. The asynchronous transfer mode (ATM), which is now widely accepted as the basis for this network [2], offers very flexible information transfer with respect to bandwidth requirements. The introduction of a special class of priorities, called space priorities, adds a quality of service flexibility to the ATM bearer service. This paper describes various space priority mechanisms and their detailed performance evaluation. Furthermore, a comparative performance study is given, indicating the excellent performance characteristics of a simple buffer management scheme called partial buffer sharing.

This paper presents a novel buffer management scheme for portable media players equipped with flash memory. Though flash memory has various advantages over magnetic disks such as small and lightweight form factor, solid-state reliability, low power consumption, and shock resistance, its physical characteristics imposes several limitations. Most notably, it takes relatively long time to write data in flash memory and the data cannot be overwritten before being erased first. Since an erase operation is performed as a unit of larger block, the employed strategy for mapping logical blocks onto physical pages affects real performance of flash memory. This article suggests a flashaware buffer management scheme that reduces the number of erase operations by selecting a victim based on its page utilization rather than based on the traditional LRU policy. Our scheme effectively minimizes the number of write and erase operations in flash memory, reducing the total execution time by 17% compared to the LRU policy 1 .

R-tree based spatial join is useful because of both its superior performance and the wide spread implementation of R-trees. We present a new R-tree join method called BFRJ (Breadth-First R-tree Join). BFRJ syn- chronously traverses both R-trees in breadth- first order while processing join computation one level at a time. At each level, BFRJ creates an intermediate join index and

The design of the Georgia Tech Time Warp (GTW, version 2.0) executive for cache-coherent sharedmemory multiprocessors is described. The programmer's interface is presented. Several optimizations used to e ciently realize key functions such as event list manipulation, memory and bu er management, and message passing are discussed. An e cient algorithm for computing GVT on shared-memory multiprocessors is described. Measurements of a wireless personal communication services (PCS) network simulation indicate the GTW simulator is able to sustain performance as high as 335,000 committed events per second for this application on a 42-processor KSR-2 machine.

A wireless sensor network is constrained by computation capability, memory space, communication bandwidth, and above all, energy supply. When a critical event triggers a surge of data generated by the sensors, congestion may occur as data packets converge toward a sink. Congestion causes energy waste, throughput reduction, and information loss. However, the important problem of congestion avoidance in sensor networks is largely open. This paper proposes a congestion-avoidance scheme based on light-weight buffer management. We describe simple yet effective approaches that prevent data packets from overflowing the buffer space of the intermediate sensors. These approaches automatically adapt the sensors' forwarding rates to nearly optimal without causing congestion. We discuss how to implement buffer-based congestion avoidance with different MAC protocols.

Catchment riparian areas are considered key zones to target mitigation measures aimed at interrupting the movement of diffuse substances from agricultural land to surface waters. Hence, unfertilized buffer strips have become a widely studied and implemented "edge of field" mitigation measure assumed to provide an effective physical barrier against nitrogen (N), phosphorus (P), and sediment transfer. To ease the legislative process, these buffers are often narrow mandatory strips along streams and rivers, across different riparian soil water conditions, between bordering land uses of differing pollution burdens, and without prescribed buffer management. It would be easy to criticize such regulation for not providing the opportunity for riparian ecosystems to maximize their provision for a wider range of ecosystem goods and services. The scientific basis for judging the best course of action in designing and placing buffers to enhance their multifunctionality has slowly increased over the last five years. This collection of papers aims to add to this body of knowledge by giving examples of studies related to riparian buffer management and assessment throughout Europe. This introductory paper summarizes discussion sessions and 13 selected papers from a workshop held in Ballater, UK, highlighting research on riparian buffers brought together under the EU COST Action 869 knowledge exchange program. The themes addressed are (i) evidence of catchment-to national-scale effectiveness, (ii) ecological functioning linking terrestrial and aquatic habitats, (iii) modeling tools for assessment of effectiveness and costs, and (iv) process understanding enabling management and manipulation to enhance pollutant retention in buffers. The combined understanding led us to consider four principle key questions to challenge buffer strip research and policy.

In today's multicomputers, software overhead dominates the message-passing latency cost. We designed two multicomputer network interfaces that signif~cantiy reduce this overhead. Both support vMual-memory-mapped communication, allowing user processes to communicate without expensive buffer management and without making system calls across the protection boundary separating user processes from the operating system kerneL Here we compare the two interfaces and discuss the performance trade-offs between them.

In this paper, we propose a novel external-memory algorithm to support view-dependent simplification for datasets much larger than main memory. In the preprocessing phase, we use a new spanned sub-meshes simplification technique to build view-dependence trees I/O-efficiently, which preserves the correct edge collapsing order and thus assures the run-time image quality. We further process the resulting view-dependence trees to build the meta-node trees, which can facilitate the run-time level-of-detail rendering and is kept in disk. During run-time navigation, we keep in main memory only the portions of the meta-node trees that are necessary to render the current level of details, plus some prefetched portions that are likely to be needed in the near future. The prefetching prediction takes advantage of the nature of the run-time traversal of the meta-node trees, and is both simple and accurate. We also employ the implicit dependencies for preventing incorrect foldovers, as well as main-memory buffer management and parallel processes scheme to separate the disk accesses from the navigation operations, all in an integrated manner. The experiments show that our approach scales well with respect to the main memory size available, with encouraging preprocessing and run-time rendering speeds and without sacrificing the image quality.

We address the problem of des%ning optimal buffer management policies in shared memory switches when packets already accepted in the switch can be dropped (pushed-out). Our goal is to maximize the overall throughput, or equivalently to minimize the overall loss probability in the system. For a system with two output ports, we prove that the optimal policy is of push-out with threshold type (POT). The same result holds if the optimality criterion is the weighted sum of the port loss probabilities. For this system, we also give an approximate method for the calculation of the optimal threshold, which we conjecture to be asymptotically correct. For the N-ported system, the optimal policy is not known in general, but we show that for a symmetric system (equal traf6c on all ports) it consists of always accepting arrivals when the buffer is not full, and dropping one from the longest queue to acco11Lmod8fe the new arrival when the buffer is full. Numerical results are provided which reveal an interesting and somewhat unexpected phenomenon. While the overall improvement in loss pmbability of the optimal POT policy over the optimal coordinate-convex policy is not very significant, the loss probab~ty of an individual output port remains approximately constant as the load on the other port varies and the optimal POT policy is applied, a property not shared by the optimal coordinateconvex policy.

The commercial deployment of LTE Release 8 is gaining significant momentum all over the globe, and LTE is evolving to LTE-Advanced, which offers various new features to meet or exceed IMT-Advanced requirements. Since LTE-Advanced targets ambitious spectral efficiency and peak throughput, it poses tremendous system design challenges to operators and manufacturers, especially for mobile terminals. This article discusses modem design issues related to carrier aggregation, enhanced ICIC for HetNet, detection of eight-layer transmission, reference signals for enhanced multi-antenna support, and HARQ buffer management. We provide an overview of technical challenges and sketch the perspectives for tackling them to exploit the full benefits of the LTE-Advanced system.

Flow control mechanisms in Network-on-Chip (NoC) architectures are critical for fast packet propagation across the network and for low idling of network resources. Buffer management and allocation are fundamental tasks of each flow control scheme. Buffered flow control is the focus of this work. We consider alternative schemes (STALL/GO, T-Error, ACK/NACK) for buffer and channel bandwidth allocation in presence of pipelined switch-to-switch links. These protocols provide varying degrees of fault tolerance support, resulting in different area and power tradeoffs. Our analysis is aimed at determining the overhead of such support when running in error-free environments, which are the typical operating mode. Implementation in the ×pipes NoC architecture and functional simulation by means of a virtual platform allowed us to capture application perceived performance, thus providing guidelines for NoC designers.

The Assured Forwarding (AF) service of the IETF DiffServ architecture provides a qualitative service differentiation between classes of traffic, in the sense that a low-priority class experiences higher loss rates and higher delays than a high-priority class. However, the AF service does not quantify the difference in the service given to classes. In an effort to strengthen the service guarantees of the AF service, we propose a Quantitative Assured Forwarding service with absolute and proportional differentiation of loss, service rates, and packet delays. We present a feedback-based algorithm which enforces the desired class-level differentiation on a per-hop basis, without the need for admission control or signaling. Measurement results from a testbed of FreeBSD PC-routers on a 100 Mbps Ethernet network show the effectiveness of the proposed service, and indicate that our implementation is suitable for networks with high data rates.

With the growth in Internet access services over networks with asymmetric links such as asymmetric digital subscriber line (ADSL) and cable-based access networks, it becomes crucial to evaluate the performance of TCP/IP over systems in which the bottleneck link speed on the reverse path (i.e., the path followed by acknowledgment) is considerably slower than that on the forward path (i.e., the path followed by data packets). In this paper, we provide guidelines for designing network control mechanisms for supporting TCP/IP, the widely used Internet transport protocol, over such asymmetric networks. The key results underlying these guidelines are as follows.

Performance needs of many database applications dictate that the entire database be stored in main memory. The Dal system is a main memory storage manager designed to provide the persistence, availability and safety guarantees one typically expects from a diskresident database, while at the same time providing very high performance by virtue of being tuned to support in-memory data. User processes map the entire database into their address space and access data directly, thus avoiding expensive remote procedure calls and bu er manager interactions typical of accesses in disk-resident commercial systems available today.

Performance gains in memory have traditionally been obtained by increasing memory bus widths and speeds. The diminishing returns of such techniques have led to the proposal of an alternate architecture, the Fully-Buffered DIMM. This new standard replaces the conventional memory bus with a narrow, high-speed interface between the memory controller and the DIMMs. This paper examines how traditional DDRx based memory controller policies for scheduling and row buffer management perform on a Fully-Buffered DIMM memory architecture. The split-bus architecture used by FBDIMM systems results in an average improvement of 7% in latency and 10% in bandwidth at higher utilizations. On the other hand, at lower utilizations, the increased cost of serialization resulted in a degradation in latency and bandwidth of 25% and 10% respectively. The split-bus architecture also makes the system performance sensitive to the ratio of read and write traffic in the workload. In larger configurations, we found that the FBDIMM system performance was more sensitive to usage of the FBDIMM links than to DRAM bank availability. In general, FBDIMM performance is similar to that of DDRx systems, and provides better performance characteristics at higher utilization, making it a relatively inexpensive mechanism for scaling capacity at higher bandwidth requirements. The mechanism is also largely insensitive to scheduling policies, provided certain ground rules are obeyed.

In this paper, we present an approach to construct a built-in block-based hierarchical index structures, like Rtree, to organize data sets in one, two, or higher dimensional space and improve the query performance towards the common query types (e.g., point query, range query) on Hadoop distributed file system (HDFS). The query response time for data sets that are stored in HDFS can be significantly reduced by avoiding exhaustive search on the corresponding data sets in the presence of index structures. The basic idea is to adopt the conventional hierarchical structure to HDFS, and several issues, including index organization, index node size, buffer management, and data transfer protocol, are considered to reduce the query response time and data transfer overhead through network. Experimental evaluation demonstrates that the built-in index structure can efficiently improve query performance, and serve as cornerstones for structured or semistructured data management.

Multimedia applications demand specific support from database management systems due to the characteristics of multimedia data and their interactive usage. This includes integrated support for high-volume and time-dependent (continuous) data types like audio and video. One critical issue is to provide handling of continuous data streams including buffer management as needed for multimedia presentations. Buffer management strategies for continuous data

Modern high performance applications require efficient and scalable collective communication operations. Currently, most collective operations are implemented based on point-to-point operations. In this paper, we propose to use hardware multicast in InfiniBand to design fast and scalable broadcast operations in MPI. InfiniBand supports multicast with Unreliable Datagram (UD) transport service. This makes it hard to be directly used by an upper layer such as MPI. To bridge the semantic gap between MPI Bcast and InfiniBand hardware multicast, we have designed and implemented a substrate on top of InfiniBand which provides functionalities such as reliability, in-order delivery and large message handling. By using a sliding-window based design, we improve MPI Bcast latency by removing most of the overhead in the substrate out of the communication critical path. By using optimizations such as a new co-root based scheme and lazy ACK, we can further balance and reduce the overhead. We have also addressed many detailed design issues such as buffer management, efficient handling of out-of-order and duplicate messages, timeout and retransmission, flow control and RDMA based ACK communication.

We propose a novel approach, called parallel rollout, to solving (partially observable) Markov decision processes. Our approach generalizes the rollout algorithm of Bertsekas and Castanon (1999) by rolling out a set of multiple heuristic policies rather than a single policy. In particular, the parallel rollout approach aims at the class of problems where we have multiple heuristic policies available such that each policy performs near-optimal for a different set of system paths. Parallel rollout automatically combines the given multiple policies to create a new policy that adapts to the different system paths and improves the performance of each policy in the set. We formally prove this claim for two criteria: total expected reward and infinite horizon discounted reward. The parallel rollout approach also resolves the key issue of selecting which policy to roll out among multiple heuristic policies whose performances cannot be predicted in advance. We present two example problems to illustrate the effectiveness of the parallel rollout approach: a buffer management problem and a multiclass scheduling problem.

Several satellite systems currently in operation or under development claim to support broadband Internet applications. In these scenarios, transmission control protocol (TCP) plays a critical role. Unfortunately, when used with satellite links, TCP suffers from a number of well-known performance problems, especially for higher data rates and high altitude satellites with longer delays. In response to these difficulties, the satellite and Internet research communities have developed a large gamut of solutions ranging from architectural modifications to changes in the TCP protocol. Among these, one approach requiring minimal modifications involves splitting the TCP connection in two or more segments with one segment connecting terrestrial nodes across the satellite network. In this paper, we consider an evolution of this idea: placing a TCP proxy on board the satellite that further subdivides the end-to-end connection into separate TCP connections between ground and space. We focus upon the efficient use of buffer resources on board the satellite, while at the same time enhancing TCP performance. We evaluate two TCP protocol versions, TCP NewReno and TCP Westwood. We consider various geosynchronous earth orbit satellite scenarios, with and without the split proxy, and with different channel error conditions (random errors, shadowing, etc). Using simulation, we show that an on-board proxy provides a number of distinct advantages and can enhance throughput up to threefold for both TCP New Reno and TCP Westwood, in some scenarios, with relatively modest on-board buffering requirements. The main contributions of this paper are: the on-board split proxy concept, the buffer management strategy, the use of a realistic "urban shadowing" model in the evaluation, and the extensive comparison of the recently announced TCP Westwood with the traditional TCP New Reno.

This paper concerns the mathematical formulation and optimal solutions for the Flexible Manufacturing Systems Scheduling Problem (FMSSP) with one vehicle. This linear formulation differs from the previously published ones as it takes into account the maximum number of jobs allowed in the system, limited input/output buffer capacities, empty vehicle trips and no-move-ahead trips simultaneously. Our objective is to propose optimal solutions for small and medium-sized instances and to examine a number of commonly used assumptions and heuristics. Computational experiments are carried out on instances adapted from Bilge and Ulusoy [Bilge, Ü., Ulusoy, G., 1995. A time window approach to simultaneous scheduling of machines and material handling system in an FMS. Operations Research 43, 1058-1070] and the following heuristics are evaluated: FIFO (First In First Out) rules for input/output buffer management; and FIFO, SPT (Shortest Processing Time), STT (Shortest Travel Time) and MOQS (Maximum Outgoing Queue Size) rules concerning the vehicle. The consequences of classical assumptions are also studied: ignoring empty trips, ignoring no-move-ahead constraints, and ignoring vehicle-disjunction constraints. The numerical experiments provide a set of optimal solutions and allow to evaluate the performances of heuristic search schemes.

Epidemic, or probabilistic, multicast protocols have emerged as a viable mechanism to circumvent the scalability problems of reliable multicast protocols. However, most existing epidemic approaches use connectionless transport protocols to exchange messages and rely on the intrinsic robustness of the epidemic dissemination to mask network omissions. Unfortunately, such an approach is not networkfriendly, since the epidemic protocol makes no effort to reduce the load imposed on the network when the system is congested.

In this paper, we present a survey on management mechanisms used for ensuring the Quality-of-Services for Voice-over-IP applications. We first address the motivations of QoS management for VoIP. We then partition the system into two management planes: data plane and control plane, and describe mechanisms in each plane respectively. In data plane, we cover several important techniques including packet classifier, buffer management, scheduling, loss recovery, and error concealment. In control plane, we describe admission control, resource provisioning, traffic engineering, and connection management etc. As we will see, admission control plays a critical role in QoS for VoIP. Thus, we will discuss methodologies used in admission control in detail. Specifically, we examine parameter-based and measurement-based admission control algorithms, and their use in VoIP. Finally, we discuss limitations of current technologies and future research issues in providing QoS to VoIP applications.

Flash memory has been widely used as a storage device for consumer devices. Recently, applications using flash memory are becoming complex and diverse. One of obstacles to its wide use is the slow write performance of flash memory caused by its erase-before-write characteristic. To enhance the write performance, FTL (Flash Translation Layer) generally uses the flash log-buffer, where data is written by the out-of-place scheme. However, current log buffer-based FTL schemes show poor performance for random write requests due to the block thrashing problem and high block associativity. Recently, flash-aware buffer cache management schemes are proposed to solve the block thrashing problem. However, they cannot also eliminate the problem completely. In this paper, we propose a novel flashaware buffer cache replacement policy. The technique selects the victim page to be evicted from buffer cache considering the recent victim page sent to the flash log buffer. Our experimental results show that the proposed technique reduces the flash I/O execution time by up to 30% 1 . Index Terms -flash memory, buffer management, page replacement, flash translation layer, embedded storage.

We study an online unit-job scheduling problem arising in buffer management. Each job is specified by its release time, deadline, and a nonnegative weight. Due to overloading conditions, some jobs have to be dropped. The goal is to maximize the total weight of scheduled jobs. We present several competitive online algorithms for various versions of unit-job scheduling, as well as some lower bounds on the competitive ratios. Ô 2 1 172. Also, in the 2-uniform case, we prove a lower bound of Ô 2 1 414 for deterministic memoryless algorithms, matching a known upper bound.

On-demand streaming from a remote server through best-effort Internet poses several challenges because of network losses and variable delays. The primary technique used to improve the quality of distributed content service is replication. In the context of the Internet, Web caching is the traditional mechanism that is used. In this article we develop a new staged delivery model for a distributed architecture in which video is streamed from remote servers to edge caches where the video is buffered and then streamed to the client through a last-mile connection. The model uses a novel revolving indexed cache buffer management mechanism at the edge cache and employs selective retransmissions of lost packets between the remote and edge cache for a best-effort recovery of the losses. The new Web cache buffer management scheme includes a dynamic adjustment of cache buffer parameters based on network conditions. In addition, performance of buffer management and retransmission policies at the edge cache is modeled and assessed using a probabilistic analysis of the streaming process as well as system simulations. The influence of different endogenous control parameters on the quality of stream received by the client is studied. Calibration curves on the QoS metrics for different network conditions have been obtained using simulations. Edge cache management can be done using these calibration curves. ISPs can make use of calibration curves to set the values of the endogenous control parameters for specific QoS in real-time streaming operations based on network conditions. A methodology to benchmark transmission characteristics using real-time traffic data is developed to enable effective decision making on edge cache buffer allocation and management strategies.

While it has been shown that TCP Vegas provides better performance compared to TCP Reno, studies have identified various issues associated with the protocol. We propose modifications to the congestion avoidance mechanism of the TCP Vegas to overcome these limitations. Unlike the solutions proposed in the past, our solution, named TCP Vegas-A, is neither dependent on optimising any critical parameter values nor on the buffer management scheme implemented at the routers and hence can be implemented solely at the end host. Our simulation experiments over wired as well as over geosynchronous and lower earth orbit satellite links show that TCP Vegas-A is able to overcome several of the identified problems-it can obtain a fairer share of the network bandwidth in wired and satellite scenarios, tackle rerouting issues, rectify Vegas's bias against higher bandwidth flows and prevail over fluctuating RTT conditions of a lower earth orbit satellite link. At the same time, Vegas-A is able to preserve the unique properties of Vegas that had made it a noteworthy protocol. q

The Cambridge Research Laboratory was founded in 1987 to advance the state of the art in both core computing and human-computer interaction, and to use the knowledge so gained to support the Company's corporate objectives. We believe this is best accomplished through interconnected pursuits in technology creation, advanced systems engineering, and business development. We are actively investigating scalable computing; mobile computing; vision-based human and scene sensing; speech interaction; computer-animated synthetic persona; intelligent information appliances; and the capture, coding, storage, indexing, retrieval, decoding, and rendering of multimedia data. We recognize and embrace a technology creation model which is characterized by three major phases:

As the Internet becomes more mature, there is a realization that improving the performance of routers has the potential to substantially improve Internet performance in general. Currently, most routers forward packets in a First-In-First-Out (FIFO) order. However, the diversity of applications supported by modern IP-based networks has resulted in unpredictable packet flows, and heterogeneous network traffic. Thus, it is becoming more reasonable to consider differentiating between different types of packets, and perhaps to consider allowing packets to specify a deadline by which it must be processed. These issues have made buffer management at routers a critical issue in providing effective quality of service to the various applications that use the network. In this paper, we study an online problem in which each packet is described by its discrete arrival time, non-negative weight and discrete deadline; arriving packets are buffered for delivery and all packets have the same process...

We consider a FIFO buffer with finite storage space. An arbitrary input stream of packets arrives at the buffer, but the output stream rate is bounded, so overflows may occur. Motivated by DiffServ, we assume that each packet has value either ½ or «, for some « ½. The buffer management task is to decide which packets to drop so as to minimize the total value of lost packets, subject to the buffer space bound, and to the FIFO order of sent packets. We consider push-out buffers, where the algorithm may eject packets from anywhere in the buffer. The best lower bound on the competitive ratio of on-line algorithms for buffer management is approximately ½ ¾. In this paper we present an on-line algorithm whose competitive ratio is approximately ½ ¿¼ for the worst case «. The best previous general upper bound was about ½ .

AbstractÐThis paper presents a software-based approach to fault-tolerant routing in networks using wormhole or virtual cut-through switching. When a message encounters a faulty output link, it is removed from the network by the local router and delivered to the messaging layer of the local node's operating system. The message passing software can reroute this message, possibly along nonminimal paths. Alternatively, the message may be addressed to an intermediate node, which will forward the message to the destination. A message may encounter multiple faults and pass through multiple intermediate nodes. The proposed techniques are applicable to both obliviously and adaptively routed networks. The techniques are specifically targeted toward commercial multiprocessors where the mean time to repair (MTTR) is much smaller than the mean time between router failures (MTBF), i.e., it is sufficient to tolerate a maximum of three failures. This paper presents requirements for buffer management, deadlock freedom, and livelock freedom. Simulation results are presented to evaluate the degradation in latency and throughput as a function of the number and distribution of faults. There are several advantages of such an approach. Router designs are minimally impacted, and thus remain compact and fast. Only messages that encounter faulty components are affected, while the machine is ensured of continued operation until the faulty components can be replaced. The technique leverages existing network technology, and the concepts are portable across evolving switch and router designs. Therefore, we feel that the technique is a good candidate for incorporation into the next generation of multiprocessor networks.

infrastructure. However, in multi-hop networks, effective bandwidth decreases with increasing number of hops, mainly due to increased spatial contention. Longer hop length flows suffer from extremely low throughputs which is highly undesirable in the envisioned scenarios for mesh networks. In this paper, we show that Queue/Buffer management, at intermediate relay mesh nodes, plays an important role in limiting the performance of longer hop length flows. We propose a novel queue management algorithm for IEEE 802.11s based mesh networks that improves the performance of multihop flows by fairly sharing the available buffer at each mesh point among all the active source nodes whose flows are being forwarded. Extensive simulations reveal that our proposed scheme substantially improves the performance of multihop flows. We also identify some important design issues that should be considered for the practical deployment of such mesh networks. Nagesh S. Nandiraju et al.

In Delay Tolerant Networks (DTN) disruptions may happen frequently as end to end path is not available all the time. Thus, delays can also be extended due to its environment nature like deep space, underwater, ocean sensor networks. In manage to achieve message delivery probability in such demanding networking situations , researchers have proposed the design of store-carry-and-forward routing protocols, here a node might accumulate a message in its buffer and carry it next to for unlimited time , awaiting till a suitable forwarding opportunity acquire .Moreover, multiple message duplication into the network to increase delivery probability. This arrangement of long-standing storage and replication force a high storage overhead on network. Therefore, efficient buffer drop policies are required to resolve on buffer, which decides messages must be dropped, while node buffers are overflow. In this paper, we propose effective buffer management drop policy E-DROP for delay tolerant networks. We illustrate that conventional buffer management policy like MOFO be unsuccessful to consider all relevant information in this framework. E-DROP policy can be adjust to minimize the metrics of relayed, dropped , average latency ,overhead ratio ,hop count and to maximize the average delivery probability and buffer time. Using simulations support on an imitation mobility models shortest path map based and Map route movements, we show that our buffer management E-DROP with random message sizes drop policy performs better as the existing MOFO.

The lack of proper project planning is often cited as one of the main reasons why projects fail to be completed in time and within budget. In this paper we identify and illuminate several possible misconceptions that go round in project management periodicals and in the mindset of practicing project managers and that may hamper successful project planning. The misconceptions relate to the role of the critical path, the critical sequence (critical chain), active schedules, and the insertion of buffers in the baseline schedule as a protective mechanism against schedule distortions during project execution. The possible fallacies are illustrated using example schedules developed for an illustrative project.

We propose a dynamic sharing of radio resources in a wireless multiuser system by combining a flexible video coding scheme, namely the currently developed scalable video coding (SVC) extension of H.264/AVC with appropriate radio link buffer management for multiuser streaming services. The paper introduces the components of the proposed system and the used cross-layer interface. Simulation results verify the benefits of our approach.

As the Internet infrastructure grows to support a variety of services, its legacy protocols are being overloaded with new functions such as traffic engineering. Today, operators engineer such capabilities through clever, but manual parameter tuning. In this paper, we propose a back-end support tool for large-scale parameter configuration that is based on efficient parameter state space search techniques and on-line simulation. The framework is useful when the network protocol performance is sensitive to its parameter settings, and its performance can be reasonably modeled in simulation. In particular, our system imports the network topology, relevant protocol models and latest monitored traffic patterns into a simulation that runs on-line in a network operations center (NOC). Each simulation evaluates the network performance for a particular setting of protocol parameters. We propose an efficient large-dimensional parameter state space search technique called "recursive random search (RRS)." Each sample point chosen by RRS results in a single simulation. An important feature of this framework is its flexibility: it allows arbitrary choices in terms of the simulation engines used (e.g., ns-2, SSFnet), network protocols to be simulated (e.g., OSPF, BGP), and in the specification of the optimization objectives. We demonstrate the flexibility and relevance of this framework in three scenarios: joint tuning of the RED buffer management parameters at multiple bottlenecks, traffic engineering using OSPF link weight tuning, and outbound load-balancing of traffic at peering/transit points using BGP LOCAL_PREF parameter.

This paper deals with improving the lead-time performance of a small packaging manufacturer in the Netherlands that had already implemented several facets of Theory of Constraints (TOC) in 1997. Since then, delivery reliability and profits have increased and lead-times reduced substantially. In an attempt to achieve further lead-time improvements in 2001, the firm recently encountered some limitations of its partial TOC implementation. In this paper we focus on the modifications of its order acceptance and buffer management system that were necessary in order to obtain the desired lead-time reduction. We describe how Workload Control principles were introduced to improve the buffering approach and planning procedures the firm had been using since 1997. It shows that Workload Control provides effective tools that can be used in combination with the Drum-Buffer-Rope mechanism in order to balance the flow of work to the production floor.