Characteristics of origin-destination pair traffic in Funet

In this paper we analyze measurements gathered from a 2.5~Gbps link in the Finnish university network (Funet) in 2004. The traffic is broken down into origin-destination (OD) pair components based on source and destination IP address. We study the traffic characteristics of these components, and identify four typical representative OD pairs. For these pairs we investigate the validity of a moving IID Gaussian model. We find that the statistical properties of these OD pairs differ significantly from each other, with only some of them close to Gaussian. The OD pairs are also found to have some cross-correlation between each other, contradicting an often made assumption about OD pair independence. Furthermore, the existence of a mean-variance relation between the OD pairs is studied. We find that there is a relation between mean and variance, but for some periods of time it is rather weak.     

Required Work in the M/M/1 Queue,with Application in IP Packet Processing

We consider the ordinary M/M/1 queue with the FIFO queueing discipline. It seems that the sum of service times of the customers in the system (or the required work, as we call it briefly) is a random variable that is not considered before. In this paper we derive the equilibrium distribution of this variable. The task is not quite trivial because of the dependencies between the elapsed service time and the number of customers in the system. Our motivation for this problem comes from the performance analysis of a dynamic memory allocation scheme of a packet buffer.     

Sub-/spl mu/s switching time in silicon-on-insulator Mach-Zehnder thermooptic switch

We have demonstrated both rise and fall times below 1 /spl mu/s with 10%-90% modulation in a silicon-on-insulator thermooptical Mach-Zehnder switch. The switch is based on 9-/spl mu/m-thick and 10-/spl mu/m-wide single-mode rib waveguides. Very fast switching was achieved by using a differential control method. The switch was driven with a digital signal processor accompanied by simple electronic circuitry.     

The discrete maximal operator in metric spaces

We study the action of the so-called discrete maximal operator on Newtonian, Hölder and BMO spaces on metric measure spaces equipped with a doubling measure and a Poincaré inequality. The discrete maximal operator has better regularity properties than the standard Hardy-Littlewood maximal operator and hence is a more flexible tool in this context.     

Size- and state-aware dispatching problem with queue-specific job sizes

We consider the dispatching problem in a size- and state-aware multi-queue system with Poisson arrivals and queue-specific job sizes. By size- and state-awareness, we mean that the dispatcher knows the size of an arriving job and the remaining service times of the jobs in each queue. By queue-specific job sizes, we mean that the time to process a job may depend on the chosen server. We focus on minimizing the mean sojourn time (i.e., response time) by an MDP approach. First we derive the so-called size-aware relative values of states with respect to the sojourn time in an M/G/1 queue operating under FIFO, LIFO, SPT or SRPT disciplines. For FIFO and LIFO, the size-aware relative values turn out to be insensitive to the form of the job size distribution. The relative values are then exploited in developing efficient dispatching rules in the spirit of the first policy iteration.     

A stochastic model for a vehicle in a dial-a-ride system

We consider a stochastic model for a system which can serve n customers concurrently, and each accepted and departing customer generates a service interruption. The proposed model describes a single vehicle in a dial-a-ride transport system and is closely related to Erlang’s loss system. We give closed-form expressions for the blocking probability, the acceptance rate, and the mean sojourn time, which are all shown to be insensitive with respect to the forms of the distributions defining the workload and interruption durations.     

A Portable Liquid Scintillation Counter for General LSC and High Sensitivity Alpha-Counting Applications

Applicability of a new portable, single tube liquid scintillation (LS) spectrometer was demonstrated for general LS applications, such as wipe tests and low level alpha counting, especially radon (²²²Rn) measurements in water. Wipe tests were performed with wad sticks. They were counted conventionally in a small volume in Eppendorf tubes with less than 200µl of LS cocktail, thus minimizing costs and waste. Small volume is specially recommended because low background can be achieved without heavy lead shielding, thus maintaining portability. Typical NRC recommended LLD's were reached for ³H and other typical LS isotopes. For ²²²Rn in water a biphasic extraction system was adopted where radon is extracted from water into a water-immiscible LS cocktail. The method is sensitive because radon can be extracted from a large water volume. It was observed that common non-evaporating "safe" cocktails with di-isopropyl naphtalene solvent are convenient to use and quite suitable for extraction. Also the isotope ²²⁶Ra can be reliably measured via production of its daughter ²²²Rn. The instrument includes pulse shape electronics to perform alpha/beta separation. This is based on the fact that in LS cocktails alphas generate pulses with longer duration than betas. The alpha/beta separation can be visualized with a two dimensional graph where the x-axis represents pulse amplitude (MCA channels) and the y-axis its length. The graphical operations are all done in standard Excel/Windows environment. Due to their longer pulses, alphas have greater y-coordinates than betas with the same x-coordinate (amplitude). With this graph, one can select a region occupied only by alphas and exclude betas. The above mentioned "safe" cocktails posses good alpha/beta separation properties. Because natural background (cosmic-rays and environmental gammas) produces beta-like pulses, they can be stripped away giving low background for alpha counting, typically a few counts per hour for the extraction samples. The LLD for ²²²Rn was 0.1 Bq/I.     

Optimal size-based opportunistic scheduler for wireless systems

Modern wireless cellular systems are able to utilize the opportunistic scheduling gain originating from the variability in the users’ channel conditions. By favoring users with good instantaneous channel conditions, the service capacity of the system can be increased with the number of users. On the other hand, for service systems with fixed service capacity, the system performance can be optimized by utilizing the size information. Combining the advantages of size-based scheduling with opportunistic scheduling gain has proven to be a challenging task. In this paper, we consider scheduling of data traffic (finite-size elastic flows) in wireless cellular systems. Assuming that the channel conditions for different users are independent and identically distributed, we show how to optimally combine opportunistic and size-based scheduling in the transient setting with all flows available at time 0. More specifically, by utilizing the time scale separation assumption, we develop a recursive algorithm that produces the optimal long-run service rate vectors within the corresponding capacity regions. We also prove that the optimal operating policy applies the SRPT-FM principle, i.e., the shortest flow is served with the highest rate of the optimal rate vector, the second shortest with the second highest rate, etc. Moreover, we determine explicitly how to implement the optimal rate vectors in the actual time slot level opportunistic scheduler. In addition to the transient setting, we explore the dynamic case with randomly arriving flows under illustrative channel scenarios by simulations. Interestingly, the scheduling policy that is optimal for the transient setting can be improved in the dynamic case under high traffic load by applying a rate-based priority scheduler that breaks the ties based on the SRPT principle.     

Architectural issues of ATM based connectionless overlay network

Connectionless service such as SMDS, or its European equivalent CBDS, can be provided by an ATM network by building a separate connectionless overlay network (CLON) on top of the ATM bearer service. A CLON is composed of a number of connectionless servers, CLSs, which provide the routing function, connected by semipermanent virtual paths (VPs). In this paper, architectural issues related to the design of a CLON are studied. In particular, we address the question of the optimal number of CLSs and their interconnection topology, assuming the underlying ATM network and the mean connectionless traffic load are known. The objective of the design process is to meet given design criteria with minimum costs. The QoS requirements of the SMDS/CBDS service are briefly reviewed and a simple cost model for the CLON network is given. The dimensioning of the logical links is also considered. Then the design process to find the optimal solution for the whole network is outlined. A simplified procedure was applied in numerical studies of the topological optimization. In order to find the general dependence of the optimal number of CLSs on the size of the network and other network parameters, the optimization procedure was applied to a large number of sample ATM networks, generated by an algorithm which is briefly described. Results from these studies are presented and discussed.     

Cucurbit[6]uril Hyperpolarized Chemical Exchange Saturation Transfer Pulse Sequence Parameter Optimization and Detectability Limit Assessment at 3.0T

Molecular imaging is the future of personalized medicine; however, it requires effective contrast agents. Hyperpolarized chemical exchange saturation transfer (HyperCEST) can boost the signal of Hyperpolarized ¹²⁹Xe MRI and render it a molecular imaging modality of high efficiency. Cucurbit[6]uril (CB6) has been successfully employed in vivo as a contrast agent for HyperCEST MRI, however its performance in a clinical MRI scanner has yet to be optimized. In this study, MRI pulse sequence parameter optimization was first performed in CB6 solutions in phosphate-buffered saline (PBS), and subsequently in whole sterile citrated bovine blood. The performance of four different depolarization pulse shapes (sinusoidal, 3-lobe sinc (3LS), rectangular (block), and hyperbolic secant (hypsec) was optimized. The detectability limits of CB6 in a clinical 3.0T MRI scanner was assessed using the optimized pulse sequences. The 3LS depolarization pulses performed best, and demonstrated 24 % depletion in a 25 μM solution of CB6 in PBS. It performed similarly in blood. The CB6 detectability limit was found to be 100 μM in citrated bovine blood with a correspondent HyperCEST depletion of 30 % ±9 %. For the first time, the HP ¹²⁹Xe HyperCEST effect was observed in red blood cells (RBC) and had a similar strength as HyperCEST in plasma.