Generic Flow Control Research Articles

Exploiting Hidden Convexity in Queueing Networks: A Novel Approach to Optimal Control of Flows

Optimal flow control in queueing networks is a challenging problem occurring in many contexts, such as data centers, cloud computing, healthcare, revenue management, and distributed networks, etc. The traditional approach has been to adopt heuristic solutions or consider infinite-horizon fluid or diffusion approximations. Motivated by emerging techniques in Robust Optimization, we propose a framework, termed Pipeline Queues, which tracks the dynamics of a queue simultaneously in terms of its queue length and waiting time. We begin by showing that the dynamics of a traditional queueing system can be equivalently modeled using this approach. Our key contribution is the uncovering of the hidden convexity resulting from our modeling approach. This leads us to tractable optimization formulations for generic flow control problems of obtaining performance guarantees on average and quantiles of waiting time, under arbitrary arrival and service distributions with non-zero initial conditions. Our model is flexible enough to capture partial observability and uncertainty of the initial state, as well as various constraints on the control policy. We apply our approach to multiple examples from the literature and numerically illustrate their application. Finally, we implemented our model on a real dataset at a major hospital in India. Our proposed policies are near optimal and perform significantly better than present heuristics.

A generic material flow control model applied in two industrial sectors

This paper addresses the problem of generic planning and control of automated material handling systems (AMHSs). We build upon previous work to provide a proof of concept for generic control of AMHSs in different domains. We present a generic control architecture for AMHSs, and apply this architecture to a material flow model with storage and sorter systems. We set up our model to be applicable to AMHSs in two different industrial sectors: Baggage Handling and Distribution. We report on performance indicators and analyze how far we can control the two industries generically in terms of software implementation. To this end, we present an impressive degree of 84% commonality in the control software code. Moreover, we highlight deviations from the generic control and give insight to control procedures that deviate from the generic code. A generic architecture that optimally exploits synergy between the different market sectors may reduce design time and costs considerably for system suppliers acting in both industries, while finding a common ground to model AMHSs in these different sectors also forms a scientific challenge.

Model of the leaky bucket ATM generic flow control mechanism: a case study on solving large cyclic models

The authors describe and solve a Markov model of the leaky bucket ATM generic flow control mechanism. The model has a space cardinality which grows quickly with its parameters and is challenging to solve. Exploiting the cyclic nature of the model, the authors develop a methodology which allows them to efficiently solve instances of the model with 3 905 134 states and 53 869 532 transitions using 29.8 Mbyte of memory and 222 Mbyte of disc storage. The CPU utilisation is high (between 70% and 90%). The methodology is new and can be easily extended to any kind of finite cyclic Markov models.

Local and congestion-driven fairness algorithm in arbitrary topology networks

Typically, bandwidth reservation is not made for data applications. Therefore, the only way to provide minimum bandwidth guarantees to such an application is by using a fairness mechanism to regulate the access to the network and by controlling the packet loss (i.e., congestion) inside the network. There are numerous works treating fairness in ring networks, however, there are almost no such works on fairness in arbitrary topology networks. The context of this work is fairness in an arbitrary topology network, the MetaNet, which employs convergence routing, a loss-free routing technique which is a variant on deflection routing. We note that minimum bandwidth guarantee combined with loss-free routing are the desired quality-of-service (QoS) attributes for most data applications. While developing the mechanisms, we also present performance measures to assess the new access- and flow-control algorithm: i) locality and congestion-driven-only the subnetwork containing conflicting traffic streams becomes involved in the fairness regulation. Furthermore, the fairness regulation is activated only when congestion occurs. This implies that when there is no congestion, nodes can access the network immediately and freely, which is a key requirement for distributed computing. ii) Scalability-the data-structure sizes used in the algorithm are a function of the switching node degree, and use constant space control signals of two bits only (the ATM standard, for example, dedicates four bits in the header of each cell to generic flow-control). iii) Linear access time in the congested subnetwork-measured by "the maximal clique in what we call the conflict graph to which a node belongs," and a frequency which is inverse linear in this parameter (when the traffic pattern stabilizes).

An evaluation of flow control in group communication

Group communication services have been successfully used to construct applications with high availability, dependability, and real-time responsiveness requirements. Flow control techniques enable group members to manage their local buffers, which they use to temporarily store multicast updates. Despite buffer overflow being one of the main causes of process failures, flow control has not been studied much in the literature. We study different flow control techniques used in some of the group communication services and present two generic flow control techniques: a conservative and an optimistic technique. All existing flow control techniques for group communication can be classified as either conservative or optimistic. We then present discrete event simulation results that compare the effect of these two generic flow control techniques on the performance of two, different atomic multicast protocols, a positive acknowledgment protocol and a negative acknowledgment protocol, under several different operating conditions. Based on the study of differed existing flow control techniques for group communication and the results obtained from the simulation experiment, we provide some design guidelines for the design and implementation of a suitable flow control technique for a given group communication service.

International standardization of B-ISDN

International standardization of B-ISDN is currently being studied in CCITT. CCITT has made 13 recommendations on various aspects of B-ISDN, such as service, physical layer, ATM layer, ATM adaptation layer, signaling, traffic control and OAM. This paper summarizes the CCITT recommendations on B-ISDN with emphasis on the architectural aspects of ATM standards and the recent CCITT activities. Various research issues in ATM such as provision of connectionless data services, generic flow control and traffic descriptors are also discussed.

Analysis of cell clumping caused by ATM network GFC protocols

Two GFC (Generic Flow Control) protocols are currently under study within the CCITT, namely the BT/NTT proposal and the Australian proposal. In this paper, their basic operating principles are recalled. Then, since it has been recognized in earlier studies that cell delay variation, or more precisely cell clumping, has a major impact on peak cell rate control performed in the usage parameter control, a method for quantifying the magnitude of cell clumping in a cell stream is proposed. This method is subsequently applied for evaluating the magnitude of cell clumping introduced by the above GFC protocols. Although they provide for three priority levels, the analysis is carried out for one priority level only. Under the assumption that each traffic source transmits at its peak cell rate and that a cell spacing device is implemented in each terminal, simulation results show that both GFC mechanisms are similar in terms of cell clumping.

MD3Q: A distributed queueing protocol with full channel capacity re-use and guarantee of bandwidth

Abstract This paper describes an enhanced distributed queueing protocol called Dynamic Distributed Dual Queue (D3Q). This protocol incorporates destination release and request cancellation. Therefore D3Q allows reuse of channel capacity and is suitable for both bus and ring architectures. After that, Multirequest D3Q (MD3Q) is introduced, which is an enhanced version of D3Q incorporating multiple outstanding requests. This protocol version includes new features for guaranteeing bandwidth to a node and for supporting multiple priorities. MD3Q is not merely an enhancement of D3Q or DQDB but rather a generalisation for Distributing Queueing (DQ) protocols. The described methods of bandwidth re-use and guarantee of bandwidth can be generally applied to all kinds of distributed queueing protocols. By using only 4 bits for the access protocol and supporting multiple levels of priority, this protocol is very suitable for performing Generic Flow Control (GFC) at the ATM User-to-Network Interface (UNI), where it is currently being studied further in CCITT SG XVIII.

Input rate flow control for high speed communications networks using burst level feedback control

AbstractWe propose a new input rate flow control scheme wherein the credit increment rate is updated periodically as the loading status varies. Based upon the observed status of each station's burst‐level activity, the network access node distributes feedback control signaling messages to the stations. These signaling messages allow the stations to adapt their credit increment rates in accordance with system burst‐loading conditions. The Generic Flow Control (GFC) field of the ATM cell header can be used to carry information characterizing the burst‐level loading activity at the switch used in the proposed scheme. We also present queueing models to study the system performance at the network access points. For this purpose, we select a sub‐network topology which involves a network switch (such as a fast packet switch in high‐speed metropolitan or wide area networks) and a number of regulated source stations which drive the network switch. To avoid packet retransmissions due to cell losses at the access switch, each user station (or CPN) implements locally a replica of the input regulation scheme. The output traffic streams from the source stations, as regulated by the local input rate control mechanism (and adapted by the status messages), load a packet switch which is modeled as a multiple‐server queueing system. Performance curves are presented to illustrate the statistical queue‐size behavior and message delays at both the source stations and the network switch.

A local fairness algorithm for gigabit LAN's/MAN's with spatial reuse

The authors present an algorithm to provide local fairness for ring and bus networks with spatial bandwidth reuse. Spatial bandwidth reuse can significantly increase the effective throughput delivered by the network. The proposed algorithm can be applied to any dual ring or bus architecture such as MetaRing. In the dual bus configuration, when transporting ATM cells, the local fairness algorithm can be implemented using two generic flow control (GFC) bits in the ATM cell header. In the performance it is shown that this local fairness algorithm can exploit the throughput advantage offered by spatial bandwidth reuse better than a global fairness algorithm. This is accomplished because it ensures fair use of network resources among nodes that are competing for the same subset of links, while permitting free access to noncongested parts of the network. The performance advantage of the local fairness scheme is demonstrated by simulating the system under various traffic scenarios and comparing the results to that of the MetaRing SAT-based global fairness algorithm. It is also shown that under certain traffic patterns, the performance of this algorithm achieves the optimal throughput result predicted by the known Max-Min fairness definition.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Traffic control in asynchronous transfer mode

The definition of traffic descriptors is discussed, and cell delay variation (CDV) that complicates usage parameter control (UPC) is analyzed. A generic flow control protocol that facilitates UPC is described. It is based on the asynchronous transfer mode ring (ATMR) protocol. The ATMR-GFC protocol achieves the guarantee of bandwidth by a cyclic reset scheme. Bandwidth calculation methods for connection admission control are reviewed from the viewpoint of different time scales. Both a real-time and offline calculation method are discussed. A combined method taking advantage of both methods is proposed.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Multiple access for BISDN

Some possible access scenarios for broadband ISDN based on asynchronous transfer mode (ATM) switching are examined. The use of existing or emerging networks such as local area networks (LANs) or metropolitan area networks (MANs) as access networks is discussed. These networks are well suited for the data-oriented information that is seen as the first main application in the BISDN. Following the geographical extension of existing LAN applications, multimedia is generally seen to be a very important broadband application. Possible access schemes based directly on ATM cells and not restricted to a specific service are then considered. The status of the discussions in the International Consultative Committee for Telephone and Telegraph (CCITT) on generic flow control, which has a close relationship to ATM-based networks in the customer premises, is summarized.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

A guide to data communications: B-ISDN and how it works

The capabilities of the emerging broadband integrated services digital network (B-ISDN) are examined. Fundamental differences between broadband and narrowband ISDN are identified. The basics of asynchronous transfer mode (ATM), the switching technique used by BISDN networks, are discussed. Channel identification, generic flow control, using ATM, and connectionless service are considered. The first services expected to make use of ATM are briefly described. >

A multiservice GFC protocol

Generic flow control (GFC) is used to provide controlled transmissions from terminals over the BISDN user-network interface. The reason for requiring controlled transmission is to allow for the multiplexing of cells from a number of terminals connected in either a dual bus or star configuration. GFC must provide fair access for each terminal, and the meaning of this is discussed for different services. It must provide a guaranteed minimum bandwidth, meeting the requirement of fair access for constant bit-rate terminals. It must also provide fair access for variable bit-rate (VBR) terminals. It is proposed that all terminals of a given class are given the same percentage reduction of their agreed peak bandwidths during overload conditions, leading to approximately similar delay distributions within the class. GFC can then be considered as providing a given quality of service per terminal (through guaranteed bandwidth) and an additional common quality of service for the unguaranteed cells. Some simulation results are provided for the dual bus configuration.

Customer network configurations and generic flow control

This paper analyses the BISDN customer network configurations as specified in CCITT Recommendation I.413 (e.g. star, bus, ring, starred bus), assumed market needs and the impact on the ATM layer functionality at the broadband user-network interface at the SB and TB reference points. Orderly and fair access of several terminals to a shared transmission medium requires a suitable access protocol (generic flow control). Its functionality and possible implementations are discussed.