性能
We propose a new exact solution algorithm for closed multiclass product-form queueing networks that is several orders of magnitude faster and less memory consuming than established methods for multiclass models, such as the Mean Value…
Task parallelism as employed by the OpenMP task construct, although ideal for tackling irregular problems or typical producer/consumer schemes, bears some potential for performance bottlenecks if locality of data access is important, which…
The ATLAS Virtual Organization is grid's largest Virtual Organization which is currently in full production stage. Hereby a case is being made that a user working within that VO is going to face a wide spectrum of different systems, whose…
In this letter, we present a closed-form approximation of the outage probability for the multi-hop amplify-and-forward (AF) relaying systems with fixed gain in Rayleigh fading channel. The approximation is derived from the outage event for…
We report on our investigations on some technologies that can be used to build disk servers and networks of disk servers using commodity hardware and software solutions. It focuses on the performance that can be achieved by these systems…
Monitoring large clusters is a challenging problem. It is necessary to observe a large quantity of devices with a reasonably short delay between consecutive observations. The set of monitored devices may include PCs, network switches, tape…
Information-driven networks include a large category of networking systems, where network nodes are aware of information delivered and thus can not only forward data packets but may also perform information processing. In many situations,…
We report the design and implementation of a call-graph profiler for GNU Octave, a numerical computing platform. GNU Octave simplifies matrix computation for use in modeling or simulation. Our work provides a call-graph profiler, which is…
This paper investigates mobility management strategies from the point of view of their need of signalling and processing resources on the backbone network and load on the air interface. A method is proposed to model the serving network and…
We analyze opportunistic schemes for transmission scheduling from one of $n$ homogeneous queues whose channel states fluctuate independently. Considered schemes consist of the LCQ policy, which transmits from a longest connected queue in…
The universal scalability law (USL) is an analytic model used to quantify application scaling. It is universal because it subsumes Amdahl's law and Gustafson linearized scaling as special cases. Using simulation, we show: (i) that the USL…
Most generic performance tools display only system-level performance data using 2-dimensional plots or diagrams and this limits the informational detail that can be displayed. Moreover, a modern relational database system, like Oracle, can…
New information technologies provide a lot of prospects for performance improvement. One of them is "Dynamic Source Code Generation and Compilation". This article shows how this way provides high performance for engineering problems.
The universal scalability law of computational capacity is a rational function C_p = P(p)/Q(p) with P(p) a linear polynomial and Q(p) a second-degree polynomial in the number of physical processors p, that has been long used for statistical…
In this paper, we analyze asymptotic delay-throughput trade-offs in mobile ad-hoc networks comprising heterogeneous nodes with restricted mobility. We show that node spatial heterogeneity has the ability to drastically improve upon existing…
ENUM is a DNS-based protocol standard for mapping E.164 telephone numbers to Internet Uniform Resource Identifiers (URIs). It places unique requirements on the existing DNS infrastructure, such as data scalability, query throughput,…
The calculation of network reliability in a probabilistic context has long been an issue of practical and academic importance. Conventional approaches (determination of bounds, sums of disjoint products algorithms, Monte Carlo evaluations,…
This paper deals with asymptotic expressions of the Mean Time To Failure (MTTF) and higher moments for large, recursive, and non-repairable systems in the context of two-terminal reliability. Our aim is to extend the well-known results of…
We propose a set of benchmarks that specifically targets a major cause of performance degradation in high performance computing platforms: irregular access patterns. These benchmarks are meant to be used to asses the performance of…
We study the conditional sojourn time distributions of processor sharing (PS), foreground background processor sharing (FBPS) and shortest remaining processing time first (SRPT) scheduling disciplines on an event where the job size of a…