Load balancing plays a critical role in efficiently dispatching jobs in parallel-server systems such as cloud networks and data centers. A fundamental challenge in the design of load balancing algorithms is to achieve an optimal trade-off between delay performance and implementation overhead (e.g. communication or memory usage). This trade-off has primarily been studied so far from the angle of the

We explore formal approximation techniques for Markov chains based on state–space reduction that aim at improving the scalability of the analysis, while providing formal bounds on the approximation error. We first present a comprehensive survey of existing state-reduction techniques based on clustering or truncation. Then, we extend existing frameworks for aggregation-based analysis of Markov chains

A distinctive feature of cloud computing is that it enables customers to dynamically summon server instances. Service providers facing uncertain demand patterns may exploit this feature by setting automatic provisioning rules for right-sizing the capacity contracted from the cloud. This situation can be modeled by a queueing system where the numbers of both jobs and servers evolve in time, the latter

We study the problem of scheduling packet transmissions with the aim of minimizing the energy consumption and data transmission delay of users in a wireless network in which spatial reuse of spectrum is employed. We approach this problem using the theory of Whittle index for cost minimizing restless bandits, which has been used to effectively solve problems in a variety of applications. We design two

We investigate the impact of random deviations in planned travel times using an extension of Vickrey’s celebrated bottleneck model. The model is motivated by the fact that in real life, users can neither exactly plan the time at which they depart from home, nor the delay they experience before they join a particular congestion bottleneck under investigation. We show that the strategy advocated by the

Streaming video continues to experience unprecedented growth. This underscores the need to identify user-centric performance measures and models that will allow operators to satisfy requirements for cost-effective network dimensioning delivered with an acceptable level of user experience. This paper presents an analysis of two novel metrics in the context of fixed capacity links: (i) the average proportion

In an input-queued switch, a crossbar schedule, or a matching between the input ports and the output ports needs to be computed in each switching cycle, or time slot. Designing switching algorithms with very low computational complexity, that lead to high throughput and small delay is a challenging problem. There appears to be a fundamental tradeoff between the computational complexity of the switching

In this work, we consider a random access Internet of Things IoT wireless network assisted by two aggregators collecting information from two disjoint groups of sensors. The nodes and the aggregators are transmitting in a random access manner under slotted time, the aggregators perform network-level cooperation for the data collection. The aggregators are equipped with queues to store data packets

We investigate the stability condition for redundancy-d systems where each of the servers follows a processor-sharing (PS) discipline. We allow for generally distributed job sizes, with possible dependence among the d replica sizes being governed by an arbitrary joint distribution. We establish that for homogeneous servers the stability condition for the associated fluid-limit model is characterized

Software-Defined Networks, in its essence, are the separation of the data and control planes of switching devices. The OpenFlow (OF) protocol is the most popular SDN protocol today, being available in many switches. This is due to the low implementation cost as well as the potential for innovative solutions in the network. Although OF is being used in many research papers and production networks, as

This paper focuses on the generation of stochastic models for dependability and performability analysis, through mechanisms for the automatic replication of template models when identity of replicas cannot be anonymous. The major objective of this work is to support the modeler in selecting the most appropriate replication mechanism, given the characteristics of the system under analysis. To this purpose

GPU-accelerated computing is a key technology to realize high-speed inference servers using deep neural networks (DNNs). An important characteristic of GPU-based inference is that the computational efficiency, in terms of the processing speed and energy consumption, drastically increases by processing multiple jobs together in a batch. In this paper, we formulate GPU-based inference servers as a batch

This paper studies the scheduling problem in a co-located wireless network under both the deadline and power constraints. We consider a frame-based time-slotted system. The channel condition of a link remains constant within each frame but varies from frame to frame. Packets with hard deadlines arrive at the transmitters at the beginning of each frame, and will be discarded if missing their deadlines

Global stability is an important aspect associated with the analysis and performance of closed-loop systems with feedback delays. The presence of time delays in the feedback makes the system infinite-dimensional. Establishing conditions for global stability of delay models is a non-trivial task when the system is non-linear, and the feedback delays are heterogeneous. In this paper, at the level of

Ethereum is a public, permissionless blockchain, with Ether as cryptocurrency, and with Turing-complete smart contracts to implement arbitrarily complex distributed applications. Correct operation of Ethereum relies on appropriately rewarding participating nodes (called miners) for the resources used to run the blockchain. In Ethereum the Used Gas determines the reward miners receive for executing

One of the distinctive features of today’s mobile networks is the densification of the access nodes and their heterogeneity, which lead to complex, multi-tier, multi-radio access systems. Unlike previous work, which has focussed on optimal techniques for user assignment and technology selection schemes, in this paper we present a flexible analytical model for the performance evaluation and the efficient

We consider a cooperative game based on a network in which nodes represent players and the characteristic function is defined using a maximal covering by the pairs of connected nodes. Problems of this form arise in many applications such as mobile communications, patrolling, logistics and sociology. The Owen value, which describes the significance of each node in the network, is derived. We show that

Offloading heavy computations from a mobile device to cloud servers can reduce the power consumption of the mobile device and improve the response time of the mobile application. However, the gains of offloading can be significantly affected by failures of cloud servers and network links. In this paper, we propose a fault-aware, multi-site computation offloading model capable of finding a near-optimal

By delegating path control to end-hosts, future Internet architectures offer flexibility for path selection. However, there is a concern that the distributed routing decisions by end-hosts, in particular load-adaptive routing, can lead to oscillations if path selection is performed without coordination or accurate load information. Prior research has addressed this problem by devising path-selection

We study an entanglement distribution switch that serves k users in a star topology. The function of the switch is to facilitate end-to-end bipartite entangled state generation for pairs of users. We study a simple variant of this problem, wherein all links connecting the users to the switch are identical, the effects of state decoherence are negligible, and the switch can store an arbitrary number

We analyse a data-processing system with n clients producing jobs which are processed in batches by m parallel servers; the system throughput critically depends on the batch size and a corresponding sub-additive speedup function. In practice, throughput optimization relies on numerical searches for the optimal batch size, a process that can take up to multiple days in existing commercial systems. In

In this paper, we propose a class of approximation algorithms for the max-weight matching (MWM) policy for input-queued switches, called expected 1-APRX. We establish the state space collapse (SSC) result for expected 1-APRX, and characterize its queue length behavior in the heavy-traffic limit. Our results indicate that expected 1-APRX can approximately approach the optimal queue length scaling in

Bitcoin-NG is among the first blockchain protocols to approach the near-optimal throughput by decoupling blockchain operation into two planes: leader election and transaction serialization. Its decoupling idea has inspired a new generation of high-performance blockchain protocols. However, the existing incentive analysis of Bitcoin-NG has several limitations. First, the impact of network capacity is

Speed scaling for a network of servers represented by a directed acyclic graph is considered. Jobs arrive at a source server, with a specified destination server, and are defined to be complete once they are processed by all servers on any feasible path between the source and the corresponding destination. Each server has variable speed, with power consumption function P, a convex increasing function

Modern data centers serve workloads which are capable of exploiting parallelism. When a job parallelizes across multiple servers it will complete more quickly, but jobs receive diminishing returns from being allocated additional servers. Because allocating multiple servers to a single job is inefficient, it is unclear how best to allocate a fixed number of servers between many parallelizable jobs.

Poisson’s equation plays a fundamental role as a tool for performance evaluation and optimization of Markov chains. For continuous-time birth–death chains with possibly unbounded transition and cost rates as addressed herein, when analytical solutions are unavailable its numerical solution can in theory be obtained by a simple forward recurrence. Yet, this may suffer from numerical instability, which

Although NAND flash memory has revolutionized how we manage data in modern digital systems, significant improvements are needed in flash-based storage systems to meet the requirements of emerging data-intensive applications. In this paper, we address the problem of NAND aging markers that represent the wearing degree of NAND cells. Since all flash operations are affected by the wearing status of NAND

We investigate in this paper uplink transmission schemes for 5G Ultra-Reliable Low Latency Communications (URLLC) traffic. The URLLC class of services has been defined for applications requiring extremely stringent latency and reliability. We show that, in systems with episodic traffic and many users compared with the number of transmission resources, randomly transmitting multiple copies of a packet

In this paper, we develop a new type of delay cycle analysis for single-class finite-buffer M/G/1 queues, and then analyze M/G/1 priority queues with a finite buffer for high-priority customers and an infinite buffer for low-priority customers using the proposed delay cycle analysis. For this type of priority queues, we derive the waiting time distributions of high- and low-priority customers, as well

We consider caching networks in which the routing cost for serving a content request can be reduced by caching the requested content item in cache nodes closer to the users. We refer to the cost reduction enabled by caching as the caching gain, and the product of the caching gain of a content request and its request rate as caching gain rate. We aim to study fair content allocation strategies through

With the increasing popularity and significance of content delivery services, especially video streaming, stringent Quality of Service (QoS) requirements have been placed upon Internet content providers (CPs). As a result, CPs have strong incentives to pay for resources such as premium peering bandwidth and cache capacity that ameliorate service quality. In this paper, we study an Internet access market

In this paper, we study a variant of PS+PS multilevel scheduling, which we call the PS+IS queue. Specifically, we use Processor Sharing (PS) at both queues, but with linear frequency scaling on the second queue, so that the latter behaves like an Infinite Server (IS) queue. The goals of the system are low response times for small jobs in the first queue, and reduced power consumption for large jobs

We consider scheduling to minimize mean response time of the M/G/k queue with unknown job sizes. In the single-server k=1 case, the optimal policy is the Gittins policy, but it is not known whether Gittins or any other policy is optimal in the multiserver case. Exactly analyzing the M/G/k under any scheduling policy is intractable, and Gittins is a particularly complicated policy that is hard to analyze

We consider the load balancing problem in large-scale heterogeneous systems with multiple dispatchers. We introduce a general framework called Local-Estimation-Driven (LED). Under this framework, each dispatcher keeps local (possibly outdated) estimates of the queue lengths for all the servers, and the dispatching decision is made purely based on these local estimates. The local estimates are updated

Strategically prefetching data has been utilized in practice to improve caching performance. Apart from caching data items upon requests, they can be prefetched into the cache before requests actually occur. The caching and prefetching operations compete for the limited cache space, whose size is typically much smaller than the number of data items. A key question is how to design an optimal prefetching

Heterogeneity is becoming increasingly ubiquitous in modern large-scale computer systems. Developing good load balancing policies for systems whose resources have varying speeds is crucial in achieving low response times. Indeed, how best to dispatch jobs to servers is a classical and well-studied problem in the queueing literature. Yet the bulk of existing work on large-scale systems assumes homogeneous

Social media platforms, such as Facebook and TikTok, have triggered debates on privacy. The recent transformation of social media into an increasingly centralized service, exemplified by TikTok, only exacerbates the matter. While aggregation has been deemed an effective way to combat privacy infringement, a high degree of centralization can make aggregation ineffective. We present a randomized article-push

Revenue sharing contracts between Content Providers (CPs) and Internet Service Providers (ISPs) can act as leverage for enhancing the infrastructure of the Internet. ISPs can be incentivised to make investments in network infrastructure that improve Quality of Service (QoS) for users if attractive contracts are negotiated between them and CPs. The idea here is that part of the revenue of CPs is shared

We consider a single server queue in continuous time, in which customers must be served before some limit sojourn time of exponential distribution. Customers who are not served before this limit leave the system: they are impatient. The fact of serving customers and the fact of losing them due to impatience induce costs. The fact of holding them in the queue also induces a constant cost per customer

The Kalray MPPA2-256 processor integrates 256 processing cores and 32 management cores on a chip. These cores are grouped into clusters and clusters are connected by a high-performance network on chip (NoC). This NoC provides hardware mechanisms (ingress traffic limiters) that can be configured to offer service guarantees. This paper introduces a network calculus formulation, designed to configure

We study the Task Assignment based on Guessing Size (TAGS) policy in a parallel and homogeneous server system. The policy parameters are the number of servers h and a set of cutoffs s1

A Markovian network of two queues, with finite size batch Poisson arrivals and departures, is solved approximately, but to arbitrary accuracy, for its equilibrium state probabilities. Below a pair of thresholds on the queue lengths, a modification of the Spectral Expansion Method is used to construct a semi-product-form at all lengths of one queue in a finite lattice strip defined by the threshold

Understanding workloads and modeling their performance is important for optimizing systems and services. A useful first step towards understanding the characteristics of workloads is to analyze their inter-arrival times and service requirements. If these characteristics are found to follow certain probability distributions, then corresponding stochastic models can be employed to efficiently estimate

We consider assignment policies that allocate resources to users, where both resources and users are located on a one-dimensional line [0,∞). First, we consider unidirectional assignment policies that allocate resources only to users located to their left. We propose the Move to Right (MTR) policy, which scans from left to right assigning nearest rightmost available resource to a user, and contrast

Compiler optimization passes employ cost models to determine if a code transformation will yield performance improvements. When this assessment is inaccurate, compilers apply transformations that are not beneficial, or refrain from applying ones that would have improved the code. We analyze the accuracy of the cost models used in LLVM’s and GCC’s vectorization passes for three different instruction

Loyalty programs are important tools for sharing platforms seeking to grow supply. Online sharing platforms use loyalty programs to heavily subsidize resource providers, encouraging participation and boosting supply. As the sharing economy has evolved and competition has increased, the design of loyalty programs has begun to play a crucial role in the pursuit of maximal revenue. In this paper, we first

A finite-buffer queueing system with threshold waking and early setup policy is investigated. The arrival stream is governed by a Poisson process while service times are assumed to be generally distributed. The natural FIFO processing discipline is used. Every time when the system empties, a type-specific energy saving policy is initialized that is a mixture of the classical N-type policy and an early

In this paper, a few models for optical router nodes are considered. The stations (ports) of such a node try to transmit packets. Successful transmission of a packet of type j at station i gives a profit γij, but there is also a positive probability that such a packet is dropped, causing a penalty θij. Consider one fixed cycle (frame), in which each station is assigned some visit time. The goal is

Large and unmanaged router buffers could lead to an increase in queuing delays in the Internet, which is a serious concern for network performance and quality of service. Our focus is to conduct a performance evaluation of Compound TCP (C-TCP), in a regime where the router buffer sizes are small (i.e., independent of the bandwidth-delay product), and the queue policy is Drop-Tail. In particular, we

As a multi-class variant of the classical egalitarian processor-sharing (EPS) discipline, discriminatory processor sharing (DPS) provides a suitable paradigm to model systems where share exists to control the service access of heterogeneous jobs. Although DPS is a more fine-grained scheduling discipline than EPS, the behavior of closed queueing networks (QNs) with DPS remains poorly understood. We

Storage is performance critical for I/O intensive applications deployed in data centers. As one of the recent implementations, media-based cache (MBC) is designed to alleviate the performance bottleneck of hard disk drives (HDD). This paper is motivated by the performance analysis of MBC in HDD and a new queueing model called M/G[b]/1/K with multiple state-dependent vacations is developed to formulate

We develop a unified framework for analyzing and optimizing costs for systems of FCFS queues with batch arrivals, setup delays and a general nonlinear cost structure that includes costs associated with energy used, setup times and Quality of Service (QoS) measures. We focus on the MX/G/1 and GeoX/G/1 queues with i.i.d. service times, but our results hold also for arbitrary i.i.d. batch structures where

Stability and diversity of end-to-end routes are key properties of Internet that have a large effect on the design of networks and networked systems. As the Internet evolves and deploys new technologies, it is important to re-assess these properties in the face of the new realities. This paper evaluates the stability and diversity of Internet routes with emphasis on the impact of the widely deployed

The paper revisits the performance evaluation of caching in a Named Data Networking (NDN) router where the content store (CS) is supplemented by a pending interest table (PIT). The PIT aggregates requests for a given content that arrive within the download delay and thus brings an additional reduction in upstream bandwidth usage beyond that due to CS hits. We extend prior work on caching with non-zero

In this paper, we consider an MX/M/1/GSET-VARI queue, which is a single server queue with a variable service speed, batch arrivals and general setup times. The service speed of the server is proportional to the number of jobs in the system. Our model is motivated by power-aware servers in data centers where dynamic scaling techniques are used. In this paper, we obtain the necessary and sufficient condition

This paper investigates the performance of the numerical inverse Laplace transformation (ILT) method based on concentrated matrix exponential (CME) distributions, referred to as the CME method. The CME method does not generate overshoot and undershoot (i.e., avoids Gibbs oscillation), preserves monotonicity of functions, its accuracy is gradually improving with the order, and it is numerically stable

We consider a resource allocation problem in a geographically distributed cloud network, where the goal is to obtain the capacities of the servers across the network in order to minimize the overall cost. In this study, the system resources (servers) are subject to failures, due to occasional breakdowns or cyber attacks. As a result, the servers supply is of a stochastic nature. From an optimization

Opportunistic scheduling in wireless cellular systems utilizes random channel quality variations in time by favoring the users with good channel conditions. However, the success of such schedulers is heavily depending on the accuracy of the available information on the channel states of users. In this paper, we consider the opportunistic scheduling problem of downlink data traffic with partial channel

Distributed and iterative network utility maximization algorithms, such as the primal–dual algorithms or the network-user decomposition algorithms, often involve trajectories where the iterates may be infeasible, convergence to the optimal points of relaxed problems different from the original, or convergence to local maxima. In this paper, we highlight the three issues with iterative algorithms. We