Dynamic Allocation Policy Research Articles

Dynamic GTS Scheduling of Periodic Skippable Slots in IEEE 802.15.4 Wireless Sensor Networks

Abstract The IEEE 802.15.4 standard offers a flexible communication solution for low-rate wireless small area networks. When the beacon mode is enabled, real-time communication is possible by using the Guaranteed Time Slots (GTS) mechanism. However, the beacon-enabled mode only supports a maximum of seven GTS allocated in each superframe, preventing other nodes to benefit from the guaranteed service. This paper introduces the SIGHT (Skips In GTS scHeduling for IEEE 802.15.4 neTworks) model, which incorporates a dynamic GTS allocation policy for IEEE 802.15.4 networks based on the (m,k)-firm task model. Our approach ensures that at least m messages are allocated in a guaranteed slot, for any k consecutive superframes. It adopts a deterministic schedulability test and an on-line prioritization heuristic algorithm called Slotted DBP (Distance Based Priority) to attend requests according to this schedulability model. A simulation model has been developed, in order to compare the behaviour in overload conditions of the proposed Slotted DBP with other approaches (DWCS and FIFO).

Adaptive Channel Buffers in On-Chip Interconnection Networks— A Power and Performance Analysis

On-chip interconnection networks (OCINs) have emerged as a modular and scalable solution for wire delay constraints in deep submicron VLSI design. OCIN research has shown that the design of buffers in the router influences the energy consumption, area overhead, and overall performance of the network. In this paper, we propose a low-power low-area OCIN architecture by reducing the number of buffers within the router. To minimize the performance degradation due to the reduced buffer size, we use the already existing repeaters along the inter-router channels to double as buffers along the channel when required. At low network loads, the proposed adaptive channel buffers function as conventional repeaters, propagating the signals. At high network loads, the adaptive channel buffers function as storage elements in addition to the router buffers. The router buffers can be assigned either statically or dynamically to the incoming packets. Static allocation reserves equal buffer space partitioned among all of the incoming packets, whereas dynamic allocation reserves buffer space on a per-flit basis, enabling higher buffer occupancy. We evaluate the proposed adaptive channel buffers with both static and dynamic buffer allocation policies in the 90-nm technology node, using 8 times 8 mesh and folded torus network topologies. Simulation results using the SPLASH-2 suite benchmarks and synthetic traffic patterns show that, by reducing the router buffer size, our proposed architecture achieves nearly 40 percent savings in router buffer power, 30 percent savings in overall network power, and 41 percent savings in area, with only a marginal 1-5 percent drop in throughput under dynamic buffer allocation and about 10-20 percent drop in throughput for statically assigned buffers.

Transaction Scheduling for Web Services Computing Applications

Abstract Web services technology provides a new computing model, which greatly accelerates application processes and responds to changing business needs within and across enterprises. Increasingly, web services transactions are key additions to business processes that access multiple web services and need to string them together. However, many of the previous approaches for scheduling transactions considered a single class of transactions with real-time constraints, whereas in practice not all transactions in web services computing applications have real-time requirements. This work studies the problem of scheduling web services transactions to deal with both real-time and non-realtime constraints existing simultaneously. We present an efficient transaction scheduling policy named the dynamic allocation policy for web services computing (DAP-WS), which aims at minimizing the number of missed real-time web services transactions and maximizing the throughput of non-real-time web services transactions. The p...

Dynamic Slotting and Pricing Decisions in a Durable Product Supply Chain

We consider a supply chain in which a manufacturer sells an innovative durable product to an independent retailer over its life cycle. We assume that the product demand follows a Bass-type diffusion process, and it is determined by the market influences, retail price of the product, and shelf space allocated to it. We consider the following retailer profit optimization strategies: (i) the myopic strategy of maximizing the current-period profit and(ii) the far-sighted strategy of maximizing the life-cycle profit. We characterize the optimal dynamic shelf-space allocation and retail pricing policies for the retailer and wholesale pricing policies for the manufacturer. We also compute these policies. Surprisingly, we find that the manufacturer, and sometimes even the retailer, is better of with a myopic retailer strategy in some cases.

Dynamic bandwidth allocation for 3G wireless systems—A fuzzy approach

3G Wireless systems are to support multiple classes of traffic with widely different characteristics and quality of service (QoS) requirements. A major challenge in this system is to guarantee the promised QoS for the admitted users, while maximizing the resource allocation through dynamic resource sharing. In the case of multimedia call, each of the services has its own distinct QoS requirements concerning probability of blocking ( P B), service access delay (SAD), and access delay variation (ADV). The 3G wireless system attempts to deliver the required QoS by allocating appropriate resources (e.g. bandwidth, buffers), and bandwidth allocation is a key in achieving this. Dynamic bandwidth allocation policies reported so far in the literature deal with audio source only. They do not consider QoS requirements. In this work, a fuzzy logic (FL)-based dynamic bandwidth allocation algorithm for multimedia services with multiple QoS ( P B, SAD, ADV, and the arrival rate) requirements are presented and analyzed. Here, each service can declare a range of acceptable QoS levels (e.g. high, medium, and low). As QoS demand varies, the proposed algorithm allocates the best possible bandwidth to each of the services. This maximizes the utilization and fair distribution of resources. The proposed allocation method is validated in a variety of scenarios. The results show that the required QoS can be obtained by appropriately tuning the fuzzy logic controller (FLC).

Efficient Dynamic Allocation with Uncertain Valuations

In this paper we consider the problem of eciently allocating a given resource or object repeatedly over time. The agents, who may temporarily receive access to the resource, learn more about its value through its use. When the agents’ beliefs about their valuations at any given time are public information, this problem reduces to the classic multi-armed bandit problem, the solution to which is obtained by determining a Gittins index for every agent. In the setting we study, agents observe their valuations privately, and the ecient dynamic resource allocation problem under asymmetric information becomes a problem of truthfully eliciting every agent’s Gittins index. We introduce two “bounding mechanisms,” under which agents announce types corresponding to Gittins indices either at least as high as or at most as high as their true Gittins indices. Using an announcement-contingent ane combination of the bounding mechanisms it is possible to implement the ecient dynamic allocation policy. We provide necessary and sucient conditions for global Bayesian incentive compatibility, guaranteeing a truthful ecient allocation of the resource. Using

Multi-threshold bandwidth reservation scheme of an integrated voice/data wireless network

In supporting various integrated services which demand different bandwidth and quality of service (QoS) requirements in these wireless networks, resource provisioning becomes a major issue. Due to the limited nature of the bandwidth of wireless links, it is essential to design wireless mobile networks that furnish effective and dynamic bandwidth allocation policies which not only have to ensure that the system guarantee potentially different QoS requirements and provide the necessary service differentiation from diverse applications, but simultaneously has to fully utilize the scarce wireless bandwidth available. To achieve this goal, we propose a generalized analytical model to investigate the performance of an integrated voice/data wireless mobile network in terms of various performance metrics which include, but are not limited to, voice-call blocking probability, forced termination probability, total channel utilization, and total carried traffic. A complete sharing system with multiple thresholds based on the guard channel policy is proposed and then analyzed using a two-dimensional Markov process.

Optimal insensitive routing and bandwidth sharing in simple data networks

Many communication systems can be efficiently modelled using queueing networks with a stationary distribution that is insensitive to detailed traffic characteristics and depends on arrival rates and mean service requirements only. This robustness enables simple engineering rules and is thus of considerable practical interest. In this paper we extend previous results by relaxing the usual assumption of static routing and balanced service rates to account for both dynamic capacity allocation and dynamic load balancing. This relaxation is necessary to model systems like grid computing, for instance. Our results identify joint dynamic allocation and routing policies for single input reversible networks that are optimal for a wide range of performance metrics. A simple two-pass algorithm is presented for finding the optimal policy. The derived analytical results are applied in a number of simple numerical examples that illustrate their modelling potential.

Stability of quantized control systems under dynamic bit assignment

In recent years, there have been several papers characterizing the minimum number of quantization levels required to assure closed-loop stability. This minimum bit rate is usually achieved through time-varying quantization policies. Many networks, however, prefer a constant bit rate configuration, so it is useful to characterize the stability of quantized feedback systems under constant bit rate quantization. This note first derives a lower bound on the number of quantization levels required for closed-loop stability under constant bit rates. We then introduce a novel dynamic bit allocation policy that achieves this bound.

A Dynamic Model of Active Portfolio Management and Mutual Fund Performance Evaluation

We analyze an optimal dynamic portfolio and asset allocation policy for investors who are concerned with the performances of their portfolios relative to a benchmark. Assuming that asset returns follow a multi-linear factor model similar to the structure of Ross (1976) and that portfolio managers adopt a mean tracking error analysis similar to Roll (1992), we develop a dynamic model of active portfolio management maximizing risk adjusted excess return over a well-diversified benchmark. Unlike the case of constant proportional portfolios for the standard utility maximization, our optimal portfolio policy is state dependent, namely a function of time to investment horizon, the return on the benchmark portfolio, and the return on the investment portfolio itself. Based on the analysis in this paper, we define a dynamic performance measure which relates portfolio's return to its risk sensitivity. Abnormal returns at each point in time are quantified as the difference between the realized and the model-fitted returns. Risk sensitivity is estimated through a dynamic matching that minimizes the total fitted error of portfolio returns. We study portfolio performances for a sample of U.S. mutual funds with the data from January 2001 to December 2003. To limit biases in the selection of a benchmark for portfolio evaluation, we assume that the benchmark portfolio is the minimum variance portfolio composed of the Dow Jones Industrial Average index and the Nasdaq 100 index components. We find that majority of the mutual funds have substantially under-performed the chosen benchmark. Our model also implies an interesting relationship between performance indices and risk sensitivities. For the three year data, the empirical analysis shows that portfolio performance indices are related to their estimated risk sensitivities in an open-upward quadratic curve.

Dynamic balancing of inventory in supply chains

We consider the dynamic version of the classic problem of allocation of inventories to a set of retailers to rectify the imbalance of inventories amongst them. While most research is focussed on analyzing different allocation strategies with a predetermined time of shipment (static policy), we investigate the benefit of using real time demand (inventory) information to schedule rebalancing shipments in a retail network. We model the dynamic rebalancing problem that has two decisions, the timing of the balancing shipments and determination of the new stocking levels at the retailers, as a dynamic program (DP). We obtain structural properties for the optimal allocation, rebalancing and timing strategies. We also present conditions under which a greedy heuristic to decide how much to ship from one retailer to another is optimal. The DP for determining the optimal timing and quantity of shipments has a very large state space. We present an algorithm to solve this DP efficiently. We also provide a heuristic solution procedure to the dynamic problem that performs very close to optimal. Numerical results show that dynamic allocation policies can lead to substantial benefits over the static policy especially in systems in which the starting inventories at the retailers are balanced or when high service levels are required.

Page Migration with Dynamic Space-Sharing Scheduling Policies: The Case of the SGI O2000

In this paper, we claim that memory migration mechanism is a useful approach to improve the execution of parallel applications in dynamic execution environments, but that their performance depends on related system components such as the processor scheduling. To show that, we evaluate the automatic memory migration mechanism provided by IRIX in Origin systems, under different dynamic processor allocation policies when executing OpenMP parallel multiprogrammed Workloads. We have focused the evaluation on the effects of the page migration mechanism on the CPU time consumed by each application, the processor allocation received, and the speedup. Results demonstrate that, if the processor scheduler is memory conscious, that is it maintains as much as possible the system stable, the automarie memory page migration mechanism provided by IRIX improves the CPU time consumed by OpenMP applications.

Implementing a dynamic processor allocation policy for multiprogrammed parallel applications in the SolarisTM

AbstractParallel applications typically do not perform well in a multiprogrammed environment that uses time‐sharing to allocate processor resources to the applications' parallel threads. Co‐scheduling related parallel threads, or statically partitioning the system, often can reduce the applications' execution times, but at the expense of reducing the overall system utilization. To address this problem, there has been increasing interest in dynamically allocating processors to applications based on their resource demands and the dynamically varying system load. The Loop‐Level Process Control (LLPC) policy (Yue K, Lilja D. Efficient execution of parallel applications in multiprogrammed multiprocessor systems. 10th International Parallel Processing Symposium, 1996; 448–456) dynamically adjusts the number of threads an application is allowed to execute based on the application's available parallelism and the overall system load. This study demonstrates the feasibility of incorporating the LLPC strategy into an existing commercial operating system and parallelizing compiler and provides further evidence of the performance improvement that is possible using this dynamic allocation strategy. In this implementation, applications are automatically parallelized and enhanced with the appropriate LLPC hooks so that each application interacts with the modified version of the Solaris operating system. The parallelism of the applications are then dynamically adjusted automatically when they are executed in a multiprogrammed environment so that all applications obtain a fair share of the total processing resources. Copyright © 2001 John Wiley & Sons, Ltd.

A centralized TDMA-based scheme for fair bandwidth allocation in wireless IP networks

A unified time-division multiple access (TDMA) based centralized wireless access scheme is proposed for performing the statistical multiplexing of bursty data sources in a wireless packet data network. This scheme combines dynamic bandwidth allocation with admission control and packet conditioning (at the mobile stations) to provide fair bandwidth distribution among bursty data flows with different profile rates (or subscription levels) in an error-prone environment. The dynamic bandwidth allocation policy is credit-based and both the burst-level and the packet-level bandwidth allocations are considered. The performance of the scheme is evaluated using computer simulations for different total subscription levels, for different compositions of flows with different profile rates, and for different channel quality with different channel-error correlation patterns. The simulation results show that the throughput variability among flows with the same level of subscription is considerably small except for long range dependent flows with very high traffic burstiness. The relative throughput fairness among flows with different profile rates can also be achieved. The post facto loss and delay values (i.e., observed average packet delay and average packet loss values) for the flows depend on the corresponding delay tolerance limits of the data bursts, TDMA frame-length, and the wireless link utilization level. The energy efficiency of the wireless access scheme is evaluated in terms of the average transmitter usage time and the average receiver usage time in the mobile stations for both the burst-level and the packet-level bandwidth allocation. The proposed scheme can be used in an adaptive quality-of-service (QoS) framework for dynamically adjusting the QoS for flows in order to accommodate wireless channel errors and user mobility.

Secure buffering in firm real-time database systems

Many real-time database applications arise in electronic financial services, safety-critical installations and military systems where enforcing is crucial to the success of the enterprise. We investigate here the performance implications, in terms of killed transactions, of guaranteeing multi-level secrecy in a real-time database system supporting applications with firm deadlines. In particular, we focus on the buffer management aspects of this issue.Our main contributions are the following. First, we identify the importance and difficulties of providing secure buffer management in the real-time database environment. Second, we present , a novel buffer management algorithm that provides covert-channel-free security. SABRE employs a fully dynamic one-copy allocation policy for efficient usage of buffer resources. It also incorporates several optimizations for reducing the overall number of killed transactions and for decreasing the unfairness in the distribution of killed transactions across security levels. Third, using a detailed simulation model, the real-time performance of SABRE is evaluated against unsecure conventional and real-time buffer management policies for a variety of security-classified transaction workloads and system configurations. Our experiments show that SABRE provides security with only a modest drop in real-time performance. Finally, we evaluate SABRE's performance when augmented with the GUARD adaptive admission control policy. Our experiments show that this combination provides close to ideal fairness for real-time applications that can tolerate covert-channel bandwidths of up to one bit per second (a limit specified in military standards).

Dynamic allocation policies for the finite horizon one armed bandit problem

The unknown performance of a new experiment is to be evaluated and compared with that of an existing one over a finite horizon. The explicit structure of an optimal sequential allocation policy is obtained under pertinent reward/loss functions, when the experiments are characterized by random variables with distributions from the one parameter exponential family.

Competitive Dynamic Multiprocessor Allocation for Parallel Applications

We study dynamic multiprocessor allocation policies for parallel jobs, which allow the preemption and reallocation of processors to take place at any time. The objective is to minimize the completion time of the last job to finish executing (the makespan). We characterize a parallel job using two parameter. The job's parallelism, Pi, which is the number of tasks being executed in parallel by a job, and its execution time, li, when Pi processors are allocated to the job. The only information available to the scheduler is the parallelism of jobs. The job execution time is not known to the scheduler until the job's execution is completed. We apply the approach of competitive analysis to compare preemptive scheduling policies, and are interested in determining which policy achieves the best competitive ratio (i.e., is within the smallest constant factor of optimal). We devise an optimal competitive scheduling policy for scheduling two parallel jobs on P processors. Then, we apply the method to schedule N parallel jobs on P processors. Finally we extend our work to incorporate jobs for which the number of parallel tasks changes during execution (i.e., jobs with multiple phases of parallelism).

Using parallel program characteristics in dynamic processor allocation policies

Abstract In multiprocessors a parallel program's execution time is directly influenced by the number of processors it is allocated. The problem of scheduling parallel programs in a multiprogrammed environment becomes one of determining how to best allocate processors to the different simultaneously executing programs in order to minimize mean response time. In this paper we address the problem of how many processors to allocate to each of the executing parallel jobs by examining the following questions: 1. (1) Is allocating processors equally among all jobs (equipartitioning) a desirable property of a scheduling algorithm? 2. (2) Does using information about the service demand of parallel jobs significantly reduce mean response time? 3. (3) Does using information about the efficiency with which parallel jobs execute significantly reduce mean response time? 4. (4) Does allocating each job a number of processors corresponding to the knee of the execution time — efficiency curve significantly reduce mean response time? 5. (5) What are the desirable properties of a scheduler that is designed to minimize mean response time? The main contributions of this paper are: a first-order understanding of how processing power should be dynamically allocated to jobs, a new family of algorithms for dynamic processor allocation and a rough quantification of the benefits that may be realized by properly utilizing job characteristics when making processor allocation decisions. We believe that these new algorithms can be combined with recently demonstrated techniques for obtaining sufficiently accurate runtime estimates of job efficiencies to improve scheduler implementations for multiprogrammed multiprocessors.

Using parallel program characteristics in dynamic processor allocation policies

In multiprocessors a parallel program's execution time is directly influenced by the number of processors it is allocated. The problem of scheduling parallel programs in a multiprogrammed environment becomes one of determining how to best allocate processors to the different simultaneously executing programs in order to minimize mean response time. In this paper we address the problem of how many processors to allocate to each of the executing parallel jobs by examining the following questions: 1. (1) Is allocating processors equally among all jobs (equipartitioning) a desirable property of a scheduling algorithm? 2. (2) Does using information about the service demand of parallel jobs significantly reduce mean response time? 3. (3) Does using information about the efficiency with which parallel jobs execute significantly reduce mean response time? 4. (4) Does allocating each job a number of processors corresponding to the knee of the execution time — efficiency curve significantly reduce mean response time? 5. (5) What are the desirable properties of a scheduler that is designed to minimize mean response time? The main contributions of this paper are: a first-order understanding of how processing power should be dynamically allocated to jobs, a new family of algorithms for dynamic processor allocation and a rough quantification of the benefits that may be realized by properly utilizing job characteristics when making processor allocation decisions. We believe that these new algorithms can be combined with recently demonstrated techniques for obtaining sufficiently accurate runtime estimates of job efficiencies to improve scheduler implementations for multiprogrammed multiprocessors.

Risk-Pooling Along a Fixed Delivery Route Using a Dynamic Inventory-Allocation Policy

This paper examines static and dynamic policies for replenishing and allocating inventories amongst N retailers located along a fixed delivery route. Each retailer faces independent, normally-distributed period demand and incurs a proportional holding or backorder cost on end-of-period net-inventory. A warehouse places a system-replenishment order every m periods which is received after a fixed leadtime of T periods. Immediately upon receipt, a delivery vehicle leaves the warehouse with the system-replenishment quantity and travels to the retailers along a fixed route with fixed leadtimes between successive retailers. The warehouse holds no inventory. Under the “static” policy, allocations are determined for all retailers simultaneously at the moment the delivery vehicle leaves the warehouse. Under the “dynamic” policy, allocations are determined sequentially upon arrival of the delivery vehicle at each retailer. Our major analytical results, under appropriate dynamic (static) allocation assumptions, are: (1) optimal allocations under each policy involve bringing each retailer's “normalized-inventory” to a corresponing “normalized” system inventory; (2) optimal system replenishments are base-stock policies; (3) the minimum expected cost per cycle of dynamic (static) policy can be derived from an equivalent dynamic (static) composite retailer. Given this, we prove that the “risk-pooling incentive”—a simple measure of the benefit from adopting dynamic allocation policies—is always positive. Our simulation tests confirm that dynamic allocation policies yield lower costs than static policies, regardless of whether or not their respective allocation assumptions are valid. However, the magnitude of the cost savings is very sensitive to some system parameters.