Abstract: A hierarchical traffic management system and method (i.e., a QoS behavioral model) are disclosed herein. The system includes a classifier (30) operable to identify and classify incoming traffic streams and a queuing system. The queuing system (20) includes a plurality of queues and is operable to apply scheduling policies to the traffic streams. The queues (40) of the queuing system each include enqueue attributes configured to control a depth of the queue and dequeue attributes configured to control scheduling of the queue. The dequeue attributes include minimum bandwidth, maximum bandwidth, excess bandwidth, and priority, wherein each of the queues has one or more of the dequeue attributes defined.

Abstract: We consider a queuing system with multitype customers and nonhomogeneous flexible servers, in the heavy traffic asymptotic regime and under a complete resource pooling (CRP) condition. For the input-queued (IQ) version of such a system (with customers being queued at the system “entrance,” one queue per each type), it was shown in the work of Mandelbaum and Stolyar that a simple parsimonious Gcm scheduling rule is optimal in that it asymptotically minimizes the system customer workload and some strictly convex queuing costs. In this article, we consider a different—output-queued (OQ)—version of the model, where each arriving customer must be assigned to one of the servers immediately upon arrival. (This constraint can be interpreted as immediate routing of each customer to one of the “output queues,” one queue per each server.) Consequently, the space of controls allowed for an OQ system is a subset of that for the corresponding IQ system.We introduce the MinDrift routing rule for OQ systems (which is as simple and parsimonious as Gcm) and show that this rule, in conjunction with arbitrary work-conserving disciplines at the servers, has asymptotic optimality properties analogous to those Gcm rule has for IQ systems. A key element of the analysis is the notion of system server workload, which, in particular, majorizes customer workload. We show that (1) the MinDrift rule asymptotically minimizes server workload process among all OQ-system disciplines and (2) this minimal process matches the minimal possible customer workload process in the corresponding IQ system. As a corollary, MinDrift asymptotically minimizes customer workload among all disciplines in either the OQ or IQ system.

Abstract: Implementing fair scalable reader writer mutual exclusion for access to a critical section by a plurality of processing threads in a processing system is accomplished by creating a first queue node for a first thread on the first thread's stack, the queue node representing a request by the first thread to access the critical section; adding the first queue node to a queue pointed to by a single word reader writer mutex for the critical section, the queue representing a list of threads desiring access to the critical section, each queue node in the queue being on a stack of a thread of the plurality of processing threads; waiting until the first queue node has no preceding write requests as indicated by predecessor queue nodes on the queue; entering the critical section by the first thread; exiting the critical section by the first thread; and removing the first queue node from the queue.

Abstract: Router queues buffer packets during congestion epochs. A recent result by Appenzeller et al. showed that the size of FIFO queues can be reduced considerably without sacrificing utilization. While Appenzeller showed that link utilization is not affected, the impact of this reduction on other aspects of queue management such as fairness, is unclear. Recently, a new class of low-rate DoS attacks called shrews was shown to throttle TCP connections by causing periodic packet drops. Unfortunately, smaller buffer sizes make shrew attacks more effective and harder to detect since shrews need to overflow a smaller buffer to cause drops. In this paper, we investigate the relation between buffer size and the shrew sending rate required to cause damage. Using a simple mathematical model, we show that a relatively small increase in the buffer size over the value proposed by Appenzeller is sufficient to render the shrew attack ineffective. Intuitively, bigger buffers require the shrews to transmit at much higher rates to fill the router queue. However, by doing so, shrews are no longer low-rate attacks and can be detected by active queue management (AQM) techniques such as RED-PD. We verified our analysis through simulations showing that a moderate increase in the buffer size, coupled with an AQM mechanism is adequate to achieve high link utilization while protecting TCP flows from shrew attacks.

Abstract: The invention provides a congestion control scheme that is a delay based scheme that includes a scalable queue size and one-way queueing delay measurement to reduce network congestion. Queue size is managed by queue control, a scalable utility function, dynamic alpha tuning, and/or randomized alpha tuning. One-way queueing delay is accomplished by measuring backward queueing delay management using various methods of estimating the receiver clock period. Embodiments include estimating the receiver clock period using single sample and multiple sample periods. The system includes a method for detecting route change.

Abstract: A queuing system and method are provided for filling vacancies at electronic card tables in a card room by establishing electronic waiting lists. The queuing system comprises a plurality of electronic player interaction areas (EPIAs), a queue interface, and a server computer networked to the EPIAs and the queue interface. In use, new players are placed on one or more of the electronic waiting lists, while the EPIAs are monitored for vacancies. Upon detecting a vacant EPIA, the server computer automatically assigns the next new player on the corresponding waiting list to the vacant EPIA.

Abstract: A method and apparatus for migrating the session state of an original session from a first node to a second node, such that a process, executing in the migrated session, may enqueue and dequeue messages from a queue of messages and experience the same behavior from the queue of messages as if the process was enqueuing and dequeuing messages in the original session. Advantageously, all the benefits of migrating a session from one node to another node may be realized, even when the session contains queuing state information that describes the state of a queue of messages. Data that indicates how many messages one or more subscribers have removed from the queue of messages may also be stored to preserve proper behavior of the queue of messages for multiple subscribers when sessions are migrated.

Abstract: A desirable behavior of an Internet server is that a request's queuing delay depends on its service time in a linear fashion. Measuring the quality of service in terms of slowdown, the ratio of a request's queuing delay to its service time, provides a simple way to attain the objective. Moreover, it treats client requests equally regardless of their service time, whereas response time favors requests that need more processing resources. In this paper, we propose a proportional slowdown differentiation (PSD) service model on Internet servers. It aims to maintain prespecified slowdown ratios between different classes of client requests. To provide PSD services, we first derive a closed-form expression of the expected slowdown in an M/G/1 FCFS queuing system with a typical heavy-tailed service time distribution, the bounded Pareto distribution. Based on the closed-form expression, we design a queuing-theoretic strategy of processing-rate allocation. The rate allocation is realized by deploying a virtual server for each class. Simulation results show that the strategy can provide controllable PSD services on Internet servers. It, however, comes along with large variance and weak predictability due to the dynamics of Internet traffic. To address these issues, we design an integral feedback controller and integrate it into the queuing-theoretic strategy. Simulation results demonstrate that the integrated strategy is robust and can deliver predictable PSD services at a superior fine-grained level. We modified the Apache Web server with an implementation of the integrated processing-rate allocation strategy. Experimental results further demonstrate its effectiveness and feasibility in practice.

Abstract: A function block containing a process-cache tag for storing process-cache tags in the pre-stage of a process cache and an FIFO queue for each tag entry are installed as a traffic controller. The traffic controller stacks packet groups, identified as being from the same flow, in the same FIFO queue. Each FIFO queue records the logged state of the corresponding process queues, and when a packet arrives at an FIFO queue entry in a non-registered state, only its first packet is conveyed to a function block for processing the process-cache misses, and then it awaits registration in a process cache. Access to the process cache from the FIFO queue is implemented at the time that registration of the second and subsequent packets in the process cache are completed. This allows packets other than the first packet in the flow to always access the process cache for a cache hit.

Abstract: We focus on the online problem of queue management in networks providing differentiated services. As in DiffServ, packets are divided into two priority groups. Low priority packets are assigned the value of 1 and high priority packets are assigned the value of α > 1. The goal is to maximize the total value of packets that are sent. Restricted to FIFO queues, the packets must be sent by the order of their arrival, however we are allowed to preempt packets from the queue.Several deterministic online algorithms for this model have been presented in previous papers. Currently, the best online algorithm known for this problem has a competitive ratio of 1.304 for the worst case α [17]. In this work we consider randomized online algorithms. Our main result is an online policy that outperforms any deterministic policy and achieves a competitive ratio of 1.25, by using a single random bit. This result is lower than the deterministic lower bound of 1.281 [19]. We then derive a general lower bound for randomized algorithms of 1.197.A natural extension of this model is to assign arbitrary packet values to the input packets. Currently, the best competitive ratio achieved for a deterministic policy is 1.75 [7]. We present a randomized comparison based online policy with the same competitive ratio. Since no deterministic comparison based policy is known to have a competitive ratio better than 2, we believe that this result demonstrates the potential of using randomization to outperform deterministic policies, as in the two value model.

Abstract: The invention provides a queue management system and method for controlling the movement of a group of one or more people through a virtual queue line for a service. The system comprises registration means ( 50 ) for registering the group, the registration means comprising an information carrier ( 52 ) bearing a registration code and at least one ID tag ( 54 ) including ID details for the member(s) of the group. The registration means associates the registration code with an indication of group size and uniquely with the ID details. The system further comprises interface means ( 48 ) for enabling communications to and from the group, and a processor ( 32, 34 ) associated with the interface means and responsive to a communication from the group including a communicator address and the registration code for generating a registration record for the group representing the group size, the ID details and the communicator address. The processor is arranged to receive a communication from the group requesting access to the virtual queue and to monitor the place of the group in the queue line and then trigger a summons signal when the group approaches or reaches the head of the queue line. The interface means is responsive to the summons signal for initiating a communication to the communicator address for summoning the group to the service. Access control apparatus ( 22 ) at the service reads the at least one ID tag and compares the ID details with the registration record in order to evaluate whether access to the service should be permitted or prevented.

Abstract: A system for tracking a service request queue. The queue is electronically maintained and the system calculates current estimated wait time and tracks current place in queue. The system also initiates events based upon queue changes. Events include notifying users when service is ready or about ready. One or more radio frequency identification (RFID) label readers are used, and RFID labels are attached to or associates with signaling devices such as a pager or cell phone. A computer calculates queue changes and initiates events. One or more RFID readers are used to provide current queue status and wait time information to users based upon RFID labels that are within the range of the RFID's antenna.

Abstract: Consider a Jackson network that allows feedback and that has a single server at each queue. The queues in this network are classified as a single ‘target’ queue and the remaining ‘feeder’ queues. In this setting we develop the large deviations limit and an asymptotically efficient importance sampling estimator for the probability that the target queue overflows during its busy period, under some regularity conditions on the feeder queue-length distribution at the initiation of the target queue busy period. This importance sampling distribution is obtained as a solution to a non-linear program. We especially focus on the case where the feeder queues, at the initiation of the target queue busy period, have the steady state distribution corresponding to these instants. In this setting, we explicitly identify the importance sampling distribution when the feeder queue service rates exceed a specified threshold. We also relate our work to the existing large deviations literature to develop a perspective on successes and limitations of our results.

Abstract: Communication traffic management systems and methods are disclosed. A threshold status of a communication traffic queue of a second communication traffic management device which is capable of receiving communication traffic from a first communication traffic management device is determined, by comparing a depth of the queue to a corresponding threshold for instance. A rate of transfer of communication traffic to the second device, and possibly to particular queues of the second device from corresponding queues of the first device, is controlled based on the threshold status. Transmission of communication traffic from the queues of the second device may thus be in accordance with one communication traffic management scheme, whereas the content of each queue of the second device is controlled according to another communication traffic management scheme used by the first device to transfer communication traffic from its queues, to thereby provide for complex combined communication traffic management.

Abstract: Systems and methods provide a single reader single writer (SRSW) queue structure having entries that can be concurrently accessed in an atomic manner with a single memory access, The SRSW queues maybe combined to create more complicated queues, including multiple reader single writer (MRSW), single reader multiple writer (SRMW), and multiple reader multiple writer (MRMW) queues,

Abstract: An intelligent queue may be used to facilitate communication among a set business application programs. In one embodiment, the intelligent queue may interact with a variety of business application programs providing consistent service behaviors across a variety of data store products. The intelligent queue provides extended message storage, efficient journalling, interoperability with other intelligent queues, load balancing, once-only processing, as well as detailed message state information that tracks the status of each message from both the sender and the recipient's perspective.

Abstract: Method and apparatus to support efficient check-point and role-back operations for flow-controlled queues in network devices. The method and apparatus employ queue descriptors to manage transfer of data from corresponding queues in memory into a switch fabric. In one embodiment, each queue descriptor includes an enqueue pointer identifying a tail cell of a segment of data scheduled to be transferred from the queue, a schedule pointer identifying a head cell of the segment of data, and a commit pointer identifying a most recent cell in the segment of data to be successfully transmitted into the switch fabric. In another embodiment, the queue descriptor further includes a scheduler sequence number; and a committed sequence number that are employed in connection with transfers of data from queues containing multiple segments. The various pointers and sequence numbers are employed to facilitate efficient check-point and roll-back operations relating to unsuccessful transmissions into the switch fabric.

Abstract: A pager or other wireless communication device is used to keep participants of a queue informed of remaining wait time. An estimate of wait time is generated by a host management application using information concerning availability of a resource, for example, restaurant tables. The wait time is transmitted to each respective pager within the transmitting range of a base station. For positions high in the queue, a chit bearing an estimated return time may be issued in lieu of a pager. The chit may be returned in exchange for a pager after the return time has passed. Advertising messages may also be transmitted to the pagers, and one or more games may be played using the pagers. Pagers may be at least partially disabled when out of range of a transmitter base station.

Abstract: In a node to forward data on a switch fabric, a method that includes scheduling the forwarding of data associated with one of a plurality of traffic classes. The data is to be transmitted through one of a plurality of ports coupled to the switch fabric, each port to be associated with a queue to store data to be forwarded from that port. The scheduling is to include a two stage arbitration scheme. The first stage is to select one queue associated for each traffic class. The second stage is to select one queue from among the queues selected for each traffic class selected in the first stage.

Abstract: A system and method for work request queuing for an intelligent network interface card or adapter (402) More specifically, the invention provides a method and system that efficiently supports an extremely large number of work request queues ((406,408) A virtual queue interface is presented to the host, and supported on the 'back end' by a real queue (604) shared among many multiple virtual queues (602) A message queue subsystem for an RDMA capable network interface (402) includes a memory mapped virtual queue interface The queue interface has a large plurality of virtual message queues with each virtual queue mapped to a specified range of memory address space The subsystem includes logic to detect work requests on a host interface bus to at least one of specified address ranges corresponding to one of the virtual queues and logic to place the work requests into a real queue that is memory based and shared among at least some of the plurality of virtual queues, and wherein real queue entries include indications of the virtual queue (602) to which the work request was addressed.

Abstract: A data processing system for automatically queuing a patron. A patron registers for a queue at a merchant's place of business by using a Personal Communication Device (PCD) and is placed in a queue. Subsequent to being placed in the queue, the patron waits for an event to occur. When the event occurs, the patron is notified by way of the PCD. The patron acknowledges notification via the PCD and is removed from the queue.

Abstract: In a dynamic queue load balancing method using a plurality of computers, the load level of each computer is obtained. A queue management table including information on a queue length and the number of queued requests in each queue and a queue adjustment table including information on a queue length at each load level are referred to. The queue length in the queue management table is changed to the queue length in the queue adjustment table corresponding to the obtained load level by a storage device. A queue usage rate of queue is calculated from the queue length and the number of queued requests in the queue management table. Information on the calculated queue usage rate of queue is exchanged among the computers through a communication device. According to the exchanged information on the queue usage rate, selective assignment of a processing request among queues of the computers is performed.

Abstract: Queue spillback is a common phenomenon in congested transportation networks. Nevertheless, traditionally, dynamic traffic assignment (DTA) problems are developed with the point-queue concept in which queue spillback is not captured. Indeed, one recent focus in DTA research is to capture this phenomenon and develop solution methods for the physical- queue DTA formulations. However, the properties of these problems, which have important implications on the theoretical advances and computational issues on transportation planning and operations, are not well recognized and understood. This paper summarizes the properties of physical-queue DTA, compares those with point-queue DTA, and discusses their implications. In particular, the interrelationship among properties including First-In-First-Out, causality, travel-time-link-flow consistency, and queue spillback are emphasized in this paper.

Abstract: A method, computer program product, and a data processing system for posting and retrieving WQEs to a shared receive queue in a manner that alleviates head-of-line blocking issues is provided. The present invention provides a shared receive queue and a posting routine for allowing non-sequential work request postings in the shared receive queue. Additionally, a fetch routine is provided for non-sequential fetching of work queue elements from the shared receive queue.

Abstract: A medium access control (MAC) scheduler is disclosed for scheduling uplink (UL) traffic by a subscriber station having multiple active service connections. The scheduler may include two types of queue sets, a first type of queue for each unsolicited grant service (UGS) connection and a second type of queue for each and all other non-UGS connections. Upon receipt of an overall bandwidth grant from a base station, data in the first type of queues may be sent first and then data in the second type of queues is sent if there is sufficient remaining burst space in the granted UL frame. The second type of queues may be assigned weight value, and thus scheduled, depending on the type of connection. When serving the second type of queues, initial burst space allocation may be reserved for bandwidth requests to the base station. Additional embodiments and variations are also disclosed.

Abstract: Techniques to pace the transmission of packets to multiple connections using one or more queues or storages. When multiple queues are used, each queue may have an associated throughput rate and each of the queues may have a different associated throughput rate.