Building performance models for software services in DevOps is costly and error-prone. Accurate service demand distribution estimation critical to precisely modeling queueing behaviors prediction. However, current methods focus on capturing the mean demand, disregarding higher-order moments of that still can largely affect prediction accuracy. To address this limitation, we propose estimate hig...

Markovian Process Algebras approximate their model of synchronisation events in order to preserve their Markovian nature. This paper investigates synchronisation models in a stochastic context and focuses on how the Markovian approximation of synchronisation a ects the accuracy of the performance model. TIPP and PEPA are used as speci c cases throughout, and their di erent methods of synchronis...

This paper analyzes a finite-buffer queueing system, where customers arrive in batches and the accepted are served by single server. The service is assumed to be dependent on batch-size follows general bulk rule. inter-arrival times of correlated they represented through batch Markovian arrival process (BMAP). Computation procedure queue-length distributions at post-batch-service completion, an...

Markovian Process Algebras approximate their model of synchronisation events in order to preserve their Markovian nature. This paper investigates synchronisation models in a stochastic context and focuses on how the Markovian approximation of synchronisation a ects the accuracy of the performance model. TIPP and PEPA are used as speci c cases throughout, and their di erent methods of synchronis...

Markovian arrival processes are a powerful class of stochastic processes to represent stochastic workloads that include autocorrelation in performance or dependability modeling. However, fitting the parameters of a Markovian arrival process to given measurement data is non-trivial and most known methods focus on a single class case, where all events are of the same type and only the sequence of...

Generic non-Markovian quantum processes have infinitely long memory, implying an exact description that grows exponentially in complexity with observation time. Here, we present a finite memory ansatz approximates (or recovers) the true process errors bounded by strength of memory. The introduced is operational quantity and depends on way probed. Remarkably, recovery error smallest over all pos...

Limited preemptive scheduling has been shown to dominate both non-preemptive and fully preemptive scheduling under fixed priority systems, as far as schedulability is concerned. This paper suggests the use of DVS and DMP techniques under limited preemptive scheduling to further reduce energy consumption with respect to a fully preemptive or non-preemptive approach.