Search

Modern resource management frameworks guarantee low tail latency for long-running services using the resource over-provisioning method, resulting in serious waste of resources and increasing the service costs greatly. To reduce the over-provisioning cost, we present HRSE, a hybrid resource scaling engine that enables much more efficient resource provisioning for both periodic and non-periodic workloads of long-running services while guaranteeing the tail latency Service Level Objective (SLO). HRSE employs a convolution-based time series analysis to identify periodic patterns in workloads. If periodic patterns are discovered, HRSE estimates the just-right amount of resources based on the periodic features through a top-K based collaborative filtering approach. Otherwise, it leverages wavelet-clustering to capture the short-term patterns in non-periodic workloads and predict the resource demands for the near future. To further enforce the tail latency SLO, HRSE uses an online reprovisioning mechanism that dynamically adjusts the resources to mitigate the performance uncertainty due to workload burstinesses. We fully implement HRSE on top of Docker and conduct extensive experiments using traces from production systems. Testbed experiments show that HRSE is able to increase the average resource utilization to 43% and 45% for periodic and non-periodic workloads respectively while guaranteeing the same tail latency objective.

Edge computing is an important pillar for green computation by bringing the Cloud resources to the Edge, serving real-time applications, and reducing the computing and network resources required to transfer data for processing in the Cloud. 5G brings network densification and enables massive IoT and V2X applications, which triggers the need for edge computing to host network functions and user-facing services in a converged edge platform(s). Several edge computing deployments are being observed by ecosystem players (telco, ISVs, chip vendors, CSPs,... etc.) for IoT or V2X services, however, focusing on converged network functions and services. The point that is still in its early stages is the dynamic workload orchestration across the converged edge platforms running network functions and multi-tenant IoT services with different compute requirements and different Service Level Objectives (SLOs). This paper focuses on autonomous life cycle management for converged edge platform(s) to enable resource-efficient workload orchestration, contributing to the green goal. We present a solution for intelligent dynamic resources configuration on edge computing platforms hosting multi-tenant services while guaranteeing the SLO for each service and helping green communication goal. The presented solution has been deployed in a trial, and we present results on efficient resources configuration.

Service function chains, or more generally, virtual network services, evolve throughout their life-cycle due to traffic fluctuations, resource usage, and the stringent requirements of the SLO (Service Level Objective) that must be fulfilled. Network operators resort to elasticity mechanisms (e.g., vertical and horizontal scaling) and migration procedures to adapt the service function chains dynamically to meet their requirements constantly. However, most state-of-the-art solutions do not evaluate the overall impact of such mechanisms on the service function chains or the resulting penalty costs (e.g., migration downtime and SLO violation). To bridge this gap, we propose DAVINCI, a decision-making tool with different adaptation policies that allows the network operator to adapt the service function chains to minimize network changes (e.g., reallocating paths and VNFs) while keeping penalty costs at their lowest level. Moreover, DAVINCI allows for the adaptations (e.g., migration, vertical and horizontal scaling) to be expressed as a set of decisions and leverages the Min Cost Flow problem to estimate migration downtime. The analytical evaluation shows that DAVINCI outperforms the existing state-of-the-art solution (NFV-PEAR) in terms of migration, traffic costs, and the overhead induced by the migration and elasticity mechanisms while significantly reducing penalty costs.

Designing survivable communication networks to achieve carrier-grade five-nines reliability is of paramount importance for the network operators. This article addresses service reliability and its related aspects such as nodal reachability, network connectivity, and edge-disjoint routing in both traditional networks and software defined networks (SDNs). The proposed roadmap is based on two phases: Fundamental analytical phase and performance evaluation phase. In the first phase, a graph operator is defined to analyze the characteristics of the reliability metric and its associated reachability feature. This phase will focus on both the macro- and micro-level properties of reliability. In the second phase, we exploit the analysis in the former phase to get an insight into the performance evaluation of traditional and SDN-based networks against the reliability metric, and then calculate the statistical significance of the mean difference of their reliability values. Reliability under edge-disjoint paths to avoid resource competition is also investigated. Various types of topologies are utilized to test the service reliability of both architecture designs. Extensive simulation results show that SDN-based networks have comparable performance to its legacy counterpart against the operational reliability metric. Our findings not only shed light on enhancing reliability using edge-disjoint paths under link failure scenarios but also expected to benefit the operators to achieve their service level objectives while migrating from legacy to SDN-based platform.

Cloud computing is an important technology for businesses and individual users to obtain computing resources over the Internet on-demand and flexibly. Although cloud computing has been adopted across diverse applications, the owners of time-and-performance critical applications require cloud service providers’ guarantees about their services, such as availability and response times. Service Level Agreements (SLAs) are a mechanism to communicate and enforce such guarantees typically represented as service level objectives (SLOs), and financial penalties are imposed on SLO violations. Due to delays and inaccuracies caused by manual processing, an automatic method to periodically verify SLA terms in a transparent and trustworthy manner is fundamental to effective SLA monitoring, leading to the acceptance and credibility of such service to the customers of cloud services. This paper presents a blockchain-based distributed infrastructure that leverages fundamental blockchain properties to achieve immutable and trustworthy SLA monitoring within cloud services. The paper carries out an in-depth empirical investigation for the scalability of the proposed system in order to address the challenge of transparently enforcing real-time monitoring of cloud-hosted services leveraging blockchain technology. This will enable all the stakeholders to enforce accurate execution of SLA without any imprecisions and delays by maintaining an immutable ledger publicly across blockchain network. The experimentation takes into consideration several attributes of blockchain which are critical in achieving optimum performance. The paper also investigates key characteristics of these factors and their impact to the behaviour of the system for further scaling it up under various cases for increased service utilization.

A Service Level Agreement (SLA) regulates the provisioning of a service by defining a set of guarantees. Each guarantee sets a Service Level Objective (SLO) on some service metrics, and optionally a compensation that is applied when the SLO is unfulfilled or overfulfilled. Currently, there are software tools and research proposals that use the information about compensations to automate and optimise certain parts of the service management. However, they assume that compensations are well defined, which is too optimistic in some circumstances and can lead to undesirable situations. In this article we discuss about the notion of validity of guarantees with a compensation, which we refer to as compensable guarantees (CG). We describe an abstract model of CGs and we provide a technique that leverages constraint satisfaction problem solvers to automatically validate them. We also present a materialisation of the model of CGs in iAgree, a language to specify SLAs and a tooling support that implements our whole approach. An assessment over 319 CGs taken from 24 real-world SLAs suggests that the expressiveness and effectiveness of our proposal can pave the way for using CGs in a safer and more reliable way.

New virtualization technologies allow Infrastructure Providers (InPs) to lease their resources to Application Service Providers (ASPs) for highly scalable delivery of cloud services to end-users. However, existing literature lacks knowledge on Quality of Experience (QoE)-oriented cloud service orchestration algorithms that can guide ASPs on how to plan their budget to enhance satisfactory QoE delivery to end-users. In contrast to the InP’s cloud service orchestration, the ASP’s orchestration should not rely on expensive infrastructure control mechanisms such as Software-Defined Networking (SDN), or require apriori knowledge on the number of services to be instantiated and their anticipated placement location within InP’s infrastructure. In this paper, we address this issue of delivering satisfactory user QoE by synergistically optimizing both ASP’s management and data planes . The optimization within the ASP management plane first maximizes Service Level Objective (SLO) coverage of users when application services are being deployed, and are not yet operational. The optimization of the ASP data plane then enhances satisfactory user QoE delivery when applications services are operational with real user access. Our evaluation of QoE-oriented algorithms using realistic numerical simulations, real-world cloud testbed experiments with actual users and ASP case studies show notably improved performance over existing cloud service orchestration solutions.

More and more organizations move their data and workload to commercial cloud storage systems. However, the multiplexing and sharing of the resources in a cloud storage system present unpredictable data access latency to tenants, which may make online data-intensive applications unable to satisfy their deadline requirements. Thus, it is important for cloud storage systems to provide deadline guaranteed services. In this paper, to meet a current form of service level objective (SLO) that constrains the percentage of each tenant’s data access requests failing to meet its required deadline below a given threshold, we build a mathematical model to derive the upper bound of acceptable request arrival rate on each server. We then propose a Deadline Guaranteed storage service (called DGCloud ) that incorporates three basic algorithms. Its deadline-aware load balancing scheme redirects requests and creates replicas to release the excess load of each server beyond the derived upper bound. Its workload consolidation algorithm tries to maximally reduce servers while still satisfying the SLO to maximize the resource utilization. Its data placement optimization algorithm re-schedules the data placement to minimize the transmission cost of data replication. We further propose three enhancement methods to further improve the performance of DGCloud . A dynamic load balancing method allows an overloaded server to quickly offload its excess workload. A data request queue improvement method sets different priorities to the data responses in a server’s queue so that more requests can satisfy the SLO requirement. A wakeup server selection method selects a sleeping server that stores more popular data to wake up, which allows it to handle more data requests. Our trace-driven experiments in simulation and Amazon EC2 show the superior performance of DGCloud compared with previous methods in terms of deadline guarantees and system resource utilization, and the effectiveness of its individual algorithms.

Autoscaling methods are used for cloud-hosted applications to dynamically scale the allocated resources for guaranteeing Quality-of-Service (QoS). The public-facing application serves dynamic workloads, which contain bursts and pose challenges for autoscaling methods to ensure application performance. Existing State-of-the-art autoscaling methods are burst-oblivious to determine and provision the appropriate resources. For dynamic workloads, it is hard to detect and handle bursts online for maintaining application performance. In this article, we propose a novel burst-aware autoscaling method which detects burst in dynamic workloads using workload forecasting, resource prediction, and scaling decision making while minimizing response time service-level objectives (SLO) violations. We evaluated our approach through a trace-driven simulation, using multiple synthetic and realistic bursty workloads for containerized microservices, improving performance when comparing against existing state-of-the-art autoscaling methods. Such experiments show an increase of × 1.09 in total processed requests, a reduction of × 5.17 for SLO violations, and an increase of × 0.767 cost as compared to the baseline method. This work was partially supported by the European Research Council (ERC) under the EU Horizon 2020 programme (GA 639595), the Spanish Ministry of Economy, Industry and Competitiveness (TIN2015-65316-P and IJCI2016-27485) and the Generalitat de Catalunya (2014-SGR-1051).

We present OptEx, a closed-form model of job execution on Apache Spark, a popular parallel processing engine. To the best of our knowledge, OptEx is the first work that analytically models job completion time on Spark. The model can be used to estimate the completion time of a given Spark job on a cloud, with respect to the size of the input dataset, the number of iterations, and the number of nodes comprising the underlying cluster. Experimental results demonstrate that OptEx yields a mean relative error of 6 percent in estimating the job completion time. Furthermore, the model can be applied for estimating the cost-optimal cluster composition for running a given Spark job on a cloud under a completion deadline specified in the SLO (i.e., Service Level Objective). We show experimentally that OptEx is able to correctly estimate the required cluster composition for running a given Spark job under a given SLO deadline with an accuracy of 98 percent. We also provide a tool which can classify Spark jobs into job categories based on bisimilarity analysis on lineage graphs collected from the given jobs.

Network slicing enables communication service providers to partition physical infrastructure into logically independent networks. Network slices must be provisioned to meet the service-level objectives (SLOs) of disparate offerings, such as enhanced mobile broadband, ultrareliable low-latency communications, and massive machine-type communications. Network orchestrators must customize service placement and scaling to achieve the SLO of each network slice. In this article, we discuss the challenges encountered by network orchestrators in allocating resources to disparate 5G network slices, and propose the use of artificial intelligence to make core placement and scaling decisions that meet the requirements of network slices deployed on shared infrastructure. We explore how artificial intelligence-driven scaling algorithms, coupled with functionality-aware placement, can enable providers to design closed-loop solutions to meet the disparate SLOs of future network slices.

Nearly all principal cloud providers now provide burstable instances in their offerings. The main attraction of this type of instance is that it can boost its performance for a limited time to cope with workload variations. Although burstable instances are widely adopted, it is not clear how to efficiently manage them to avoid waste of resources. In this article, we use predictive data analytics to optimize the management of burstable instances. We design CEDULE+, a data-driven framework that enables efficient resource management for burstable cloud instances by analyzing the system workload and latency data. CEDULE+ selects the most profitable instance type to process incoming requests and controls CPU, I/O, and network usage to minimize the resource waste without violating Service Level Objectives (SLOs). CEDULE+ uses lightweight profiling and quantile regression to build a data-driven prediction model that estimates system performance for all combinations of instance type, resource type, and system workload. CEDULE+ is evaluated on Amazon EC2, and its efficiency and high accuracy are assessed through real-case scenarios. CEDULE+ predicts application latency with errors less than 10%, extends the maximum performance period of a burstable instance up to 2.4 times, and decreases deployment costs by more than 50%.

Consolidating multiple servers into a physical machine is now a commonplace in cloud infrastructures. The virtualized systems often arrange virtual disks of multiple virtual machines (VMs) on the same underlying storage device while striving to guarantee the service level objective (SLO) of the performance of each VM. Unfortunately, sync operations called by a VM may make it hard to satisfy the performance SLO by disturbing I/O activities of other VMs. In this paper, we experimentally uncover that the disk cache flush operation incurs significant I/O interference among VMs, and revisit the internal architecture and flush mechanism of the flash memory-based SSD. Then, we present vFLUSH, a novel VM-aware flush mechanism, that supports VM-based persistency control for the disk cache flush operation. We also discuss the long-tail latency issue in vFLUSH and an efficient scheme for mitigating the problem. Our evaluation with various micro- and macro-benchmarks shows that vFLUSH reduces the average latency of disk cache flush operations by up to 58.5%, thereby producing improvements in throughput by up to $1.93\times $ . The method for alleviating the long-tail latency problem, which is applied to vFLUSH, achieves a significant reduction in tail latency by up to 75.9%, with a modest throughput degradation by 2.9–7.2%.

Recent advancements in the domain of Network Function Virtualization (NFV), and rollout of next-generation networks have necessitated the requirement for the upkeep of latency-critical application architectures in future networks and communications. While Cloud service providers recognize the evolving mission-critical requirements in latency sensitive verticals such as autonomous driving, multimedia, gaming, telecommunications, and virtual reality, there is a wide gap to bridge the Quality of Service (QoS) constraints for the end-user experience. Most latency-critical services are over-provisioned on all fronts to offer reliability, which is inefficient towards scalability in the long run. To address this, we propose a strategy to model frequent violations on the application level as a multi-output target to enable more complex decision-making in the management of virtualised communication networks. In this work, we utilize data from a real-world deployment to configure and draft a realistic set of Service Level Objectives (SLOs) for a voice based NFV application, and develop a deep neural network based multi-label classification methodology to identify and predict multiple categories of SLO breaches associated with an application state. With this, we aim to gain granular SLA and SLO violation insights, enabling us to study and mitigate their impact and inform precision in drafting proactive scaling policies. We further compare the performance against a set of multi-label compatible machine learning classifiers, and address class imbalance in a multi-label setup. We perform a comprehensive evaluation to assess the performance on example-based, label-based and ranking-based measures, and demonstrate the suitability of deep learning in such a use-case.

Since the emergence of cloud computing service level objectives (SLOs) and service level agreements (SLAs) have put themselves forward as one of the key enablers for cloud's on-demand, pay-as-you-go service consumption model. To date, the vast majority of cloud platforms provide support for SLAs only in terms of statically predefined SLOs, e.g., service availability, and low-level resource capacity guarantees, e.g., CPU usage. Unfortunately, there is only limited support to clearly map workload performance requirements to the resource capacity guarantees. In this article, we introduce SLOC— a novel elasticity framework, which promotes a novel performance-driven, SLO-native approach to cloud computing. We outline the main research challenges, vision, and approach of our SLOC framework toward the SLO-native paradigm in next generation cloud computing.

Globalization has brought many changes in modern business world. Problems of enterprise-oriented software applications like distribution and configuration of resources presents challenge to the traditional software sales model. Cloud computing presents the solution for such problems. With the help of cloud-based model one can earn good return by making the QoS demands of their users (customers) satisfied. In order to fulfill the infrastructure demands (like network, storage, etc.) to their users, service providers have to maintain their own hardware or they lease it from the Iaa S providers. For maintaining their own hardware SaaS providers will have to incur an extra cost, but if taken on lease, zero maintenance cost involved. Moreover, if provider wants to optimize cost and gain customer satisfaction, they will have to satisfy their customers/users by maintaining service level objectives. This paper presents a model for resource allocation in such a way as to minimize cost and maximize profit by satisfying the service level needs of the customers. The model is designed to enable the service providers with an ability to cope with the changing needs of customers, mapping customer requests to infrastructure level parameters and better handling of heterogeneous VMs.

The power consumption of cloud data centers has a considerable impact on the environment and climate change nowadays. Researchers are seeking to find practical solutions to reduce power consumption in data centers while guaranteeing the desired level of services and service level objectives. With the establishment of the data center industry, the demand for computation and data storage has been continually rising. Energy efficiency is one of the most significant issues faced by these big data centers to meet such high computational requirements. There are many industry acceptable metrics available such as PUE, DCiE, DCP, etc. Power Usage Effectiveness (PUE) metric has proven to be the most popular in measuring energy efficiency; however, it measures the power efficiency alone with no consideration for CO2 emissions and the costs involved in total power usage across data centers. In this article, we proposed a novel Power Efficiency Measurement Calculator (PEMC) that combines and calculates the power efficiency, CO2 emissions, and the total annual costs incurred. The pseudocode and algorithm to perform these specific PUE, DCiE, and CO2 emission functions are given to explain the working of proposed work. Finally, the proposed PEMC calculator was tested and validated through a case study performed in one of the tier-level data centers in Malaysia and the results demonstrate its effectiveness compared with Power Usage Effectiveness (PUE) and other known calculators.

In the previous chapter, I introduced services and explained how they are distributed using dependency injection. When using dependency injection, the objects that are used to resolve dependencies are created by service providers, known more commonly as providers. In this chapter, I explain how providers work, describe the different types of provider that are available, and demonstrate how providers can be created in different parts of the application to change the way that services behave. Table 20-1 puts providers in context.

Despite advances in Cloud computing, ensuring high availability (HA) remains a challenge due to varying loads and the potential for Cloud outages. Deploying applications in distributed Clouds can help overcome this challenge by geo-replicating applications across multiple Cloud data centers (DCs). However, this distributed deployment can be a performance bottleneck due to network latencies between users and DCs as well as inter-DC latencies incurred during the geo-replication process. For most web applications, both HA and Performance (HAP) are essential and need to meet pre-agreed Service Level Objectives (SLOs). Efficiently placing and managing primary and backup replicas of applications in distributed Clouds to achieve HAP is a challenging task. Existing solutions consider either HA or performance but not both. In this paper we propose an approach for automating the process of providing a latency-aware failover strategy through a server placement algorithm leveraging genetic algorithms that factor in the proximity of users and inter-DC latencies. To facilitate the distributed deployment of applications and avoid the overheads of Clouds, we utilize container technologies. To evaluate our proposed approach, we conduct experiments on the Australia-wide National eResearch Collaboration Tools and Resources (NeCTAR - www.nectar.org.au) Research Cloud. Our results show at least a 23.3% and 22.6% improvement in response times under normal and failover conditions respectively compared to traditional, latency-unaware approaches. Also, the 95th percentile of response times in our approach are at most1.5 ms above the SLO compared to 11–32 ms using other approaches.

Microservice architectures are widely used in cloud-native applications as their modularity allows for independent development and deployment of components. With the many complex interactions occurring in between components, it is difficult to determine the effects of a particular microservice rollout. Site Reliability Engineers must be able to determine with confidence whether a new rollout is at fault for a concurrent or subsequent performance problem in the system so they can quickly mitigate the issue. We present Tritium, a cross-layer analytics system that synthesizes several types of data to suggest possible causes for Service Level Objective (SLO) violations in microservice applications. It uses event data to identify new version rollouts, tracing data to build a topology graph for the cluster and determine services potentially affected by the rollout, and causal impact analysis applied to metric time-series to determine if the rollout is at fault. Tritium works based on the principle that if a rollout is not responsible for a change in an upstream or neighboring SLO metric, then the rollout's telemetry data will do a poor job predicting the behavior of that SLO metric. In this paper, we experimentally demonstrate that Tritium can accurately attribute SLO violations to downstream rollouts and outline the steps necessary to fully realize Tritium.

Modern cloud services are implemented as graphs of loosely-coupled microservices to improve programmability, reliability, and scalability. Service Level Objectives (SLOs) define end-to-end latency targets for the entire service to ensure user satisfaction. In such environments, each microservice is independently deployed and (auto-)scaled. However, it is unclear how to optimally scale individual microservices when end-to-end SLOs are violated or underutilized, and how to size each microservice to meet the end-to-end SLO at minimal total cost. In this paper, we propose Parslo---a Gradient Descent-based approach to assign partial SLOs among nodes in a microservice graph under an end-to-end latency SLO. At a high level, the Parslo algorithm breaks the end-to-end SLO budget into small incremental "SLO units", and iteratively allocates one marginal SLO unit to the best candidate microservice to achieve the highest total cost savings until the entire end-to-end SLO budget is exhausted. Parslo achieves a near-optimal solution, seeking to minimize the total cost for the entire service deployment, and is applicable to general microservice graphs that comprise patterns like dynamic branching, parallel fan-out, and microservice dependencies. Parslo reduces service deployment costs by more than 6x in real microservice-based applications, compared to a state-of-the-art partial SLO assignment scheme.

Resource management for geo-distributed infrastructures is challenging due to the scarcity and non-uniformity of edge resources, as well as the high client mobility and workload surges inherent to situation awareness applications. Due to their centralized nature, state-of-the-art schedulers that work well in datacenters lack the performance and feature requirements of such applications. We present OneEdge, a hybrid control plane that enables autonomous decision-making at edge sites for localized, rapid single-site application deployment. Edge sites handle mobility, churn, and load spikes, by cooperating with a centralized controller that allows coordinated multi-site scheduling and dynamic reconfiguration. OneEdge's scheduling decisions are driven by each application's end-to-end service level objective (E2E SLO) as well as the specific requirements of situation awareness applications. OneEdge's novel distributed state management combines autonomous decision-making at the edge sites for rapid localized resource allocations with decision-making at the central controller when multi-site application deployment is needed. Using a mix of applications on multi-region Azure instances, we show that, in contrast to centralized or fully distributed control planes, OneEdge caters to the unique requirements of situation awareness applications. Compared to a centralized control plane, OneEdge reduces deployment latency by 66% for single-site applications, without compromising E2E SLOs.

5G networks are gaining pace in development and deployment in recent years. One of 5G's key objective is to support a variety of use cases with different Service Level Objectives (SLOs). Slicing is a key part of 5G that allows operators to provide a tailored set of resources to different use cases in order to meet their SLOs. Existing works focus on slicing in the frontend or the C-RAN. However, slicing is missing in the fronthaul network that connects the frontend to the C-RAN. This leads to over-provisioning in the fronthaul and the C-RAN, and also limits the scalability of the network. In this paper, we design and implement Fronthaul Slicing Architecture (FSA), which to the best of our knowledge, is the first slicing architecture for the fronthaul network. FSA runs in the switch dataplane and uses information from the wireless schedule to identify the slice of a fronthaul data packet at line-rate. It enables multipoint-to-multipoint routing as well as packet prioritization to provide multiplexing gains in the fronthaul and the C-RAN, making the system more scalable. Our testbed evaluation using scaled-up LTE traces shows that FSA can support accurate multipoint-to-multipoint routing for 80 Gbps of fronthaul traffic. Further, the slice-aware packet scheduling enabled by FSA's packet prioritization reduces the 95th percentile Flowlet Completion Times (FCT) of latency-sensitive traffic by up to 4 times.

International audience; Task scheduling and data replication are highly coupled resource management techniques that are widely used by cloud providers to improve the overall system performance and ensure service level agreement (SLA) compliance while preserving their own economic profit. However, balancing the trade-off between system performance and provider profit is very challenging. In this paper, we propose a novel scheduling algorithm called Bottleneck and Cost Value Scheduling (BCVS) algorithm coupled with a novel dynamic data replication strategy called Correlation and Economic Model-based Replication (CEMR). The main goal is to improve data access effectiveness in order to meet service level objectives in terms of response time SLORT and minimum availability SLOMA, while preserving the provider profit. The BCVS algorithm focuses on reducing system bottleneck situations caused by data transfer when the CEMR focuses on preventing future SLA violations and guaranteeing a minimum availability. An economic model is also proposed to estimate the cloud provider profit. Simulation results indicate that the proposed combination of scheduling and replication algorithms offers higher monetary profit for the cloud provider by up to 30% compared to existing strategies. Moreover, it allows better performance.

Co-location and memory sharing between latency-critical services, such as key-value store and web search, and best-effort batch jobs is an appealing approach to improving memory utilization in multi-tenant datacenter systems. However, we find that the very diverse goals of job co-location and the GNU/Linux system stack can lead to severe performance degradation of latency-critical services under memory pressure in a multi-tenant system. We address memory pressure for latency-critical services via fast memory allocation and proactive reclamation. We find that memory allocation latency dominates the overall query latency, especially under memory pressure. We analyze the default memory management mechanism provided by GNU/Linux system stack and identify the reasons why it is inefficient for latency-critical services in a multi-tenant system. We present Hermes, a fast memory allocation mechanism in user space that adaptively reserves memory for latency-critical services. It advises Linux OS to proactively reclaim memory of batch jobs. We implement Hermes in GNU C Library. Experimental result shows that Hermes reduces the average and the $99^{th}$ percentile memory allocation latency by up to 54.4% and 62.4% for a micro benchmark, respectively. For two real-world latency-critical services, Hermes reduces both the average and the $99^{th}$ percentile tail query latency by up to 40.3%. Compared to the default Glibc, jemalloc and TCMalloc, Hermes reduces Service Level Objective violation by up to 84.3% under memory pressure.

Οι τεχνολογίες πληροφοριών και επικοινωνίας (ΤΠΕ) έφεραν επανάσταση στην παροχή υπηρεσιών. Σε αυτή την εργασία, θα εστιάσουμε την προσοχή μας στις υπηρεσίες των οποίων η παροχή υποστηρίζεται από ΤΠΕ, π.χ. μέσω της χρήσης των πληροφοριακών συστημάτων των επιχειρήσεων ή μέσω του διαδικτύου. Οι υπηρεσιοστρεφείς αρχιτεκτονικές (Service Oriented Architecture – SOA) είναι σήμερα το κυρίαρχο πρότυπο για το σχεδιασμό, την ανάπτυξη και την εφαρμογή τέτοιων πληροφοριακών συστημάτων. Παραδοσιακοί κλάδοι υπηρεσιών, όπως ο τουρισμός, η υγεία και η δημόσια διοίκηση, είναι πλέον σε θέση να παρέχουν τις υπηρεσίες τους πιο αποτελεσματικά, μέσω διαφόρων διαύλων, προσαρμοσμένων στις ανάγκες των πελατών τους. Κατά συνέπεια, ένας αυξανόμενος αριθμός υπηρεσιών παρέχεται ηλεκτρονικά μέσω του διαδικτύου, διαμορφώνοντας αυτό που αναφέρεται στη βιβλιογραφία ως διαδίκτυο των υπηρεσιών (Web of Services). Η παροχή υπηρεσιών έχει προχωρήσει αναμφίβολα σε μεγάλο βαθμό. Ωστόσο, εντοπίστηκαν δύο προβλήματα που εμποδίζουν την εξατομικευμένη παροχή υπηρεσιών προσανατολισμένων στον πελάτη, ιδίως στο πλαίσιο του διαδικτύου των υπηρεσιών: i.Την περιορισμένη συμμετοχή των πελατών, ιδιαίτερα στην μοντελοποίηση και στην περιγραφή των υπηρεσιών, και ii.Την ύπαρξη πολλών ανταγωνιστικών ή επικαλυπτόμενων προτύπων και μοντέλων για υπηρεσίες, τα οποία εντέλει δεν είναι πλήρη. Επιστήμες, όπως το μάρκετινγκ και το μάνατζμεντ, ευαγγελίζονται τη σημασία της συμμετοχής του πελάτη καθ' όλη τη διάρκεια του κύκλου ζωής των υπηρεσιών και δίνουν ιδιαίτερη σημασία στην αξιολόγηση αυτών από τους πελάτες. Παρατηρήσαμε όμως ότι οι αρχές αυτές δεν τηρούνται στην παροχή υπηρεσιών στο διαδίκτυο. Οι ερευνητές στο διαδίκτυο των υπηρεσιών υποθέτουν ότι ο πελάτης έχει συνήθως έναν παθητικό ρόλο, που περιορίζεται στην κατανάλωση των υπηρεσιών και όχι απαραίτητα και στο σχεδιασμό τους. Θεωρούν ότι η μοντελοποίηση και η περιγραφή της υπηρεσίας δημιουργείται αποκλειστικά από το πάροχο. Έτσι χάνονται ή δεν εκμεταλλεύονται επαρκώς πολύτιμες πληροφορίες, γνώμες και ανατροφοδότηση που προέρχονται από τους πελάτες. Παράλληλα, διαφορετικές προσπάθειες προτυποποίησης και μοντελοποίησης των υπηρεσιών δημιούργησαν επικαλυπτόμενα, και συχνά διαφορετικά (μη-διαλειτουργικά), μοντέλα για την περιγραφή τους, κανένα από τα οποία δεν λαμβάνει υπόψη την οπτική των πελατών. Στόχος της παρούσας εργασίας είναι να αναπτυχθεί ένα ολιστικό μοντέλο δεδομένων για την υπηρεσίας που θα συνδυάζει τόσο την οπτική του παρόχου όσο και αυτή των πελατών. Πιο συγκεκριμένα, οι στόχοι της εργασίας ήταν: . Να μελετηθεί και να αναλυθεί η έννοια της υπηρεσίας, προκειμένου να δημιουργηθεί ένα κοινό βασικό μοντέλο δεδομένων. II. Να επεκταθεί το κοινό βασικό μοντέλο δεδομένων, προκειμένου να συμπεριλάβει την μοντελοποίηση της οπτικής του πελάτη για την υπηρεσία, δηλαδή πληροφορίες σχετικά με το πώς ο πελάτης την αντιλαμβάνεται και τη βιώνει. III. Να αξιολογηθεί η πληρότητα, η χρηστικότητα και η χρησιμότητα του προαναφερθέντος μοντέλου υπηρεσιών μέσω την ανάπτυξης πιλοτικών εφαρμογών που το χρησιμοποιούν προκειμένου να ικανοποιήσουν τις απαιτήσεις αναφορικά με την αναζήτηση και την εξατομίκευση υπηρεσιών στην περιοχή εφαρμογής της δημόσιας διοίκησης. Τα βασικά θεωρητικά αποτελέσματα αυτής της εργασίας περιλαμβάνουν: 1.Το μοντέλο του κύκλου ζωής της υπηρεσίας από την οπτική του πελάτη. Απεικονίζει τη συμμετοχή των πελατών κατά τις διάφορες φάσεις μιας υπηρεσίας. Περιλαμβάνει πέντε φάσεις, ήτοι: συνειδητοποίηση αναγκών, αναζήτηση, εξατομίκευση, παροχή και αξιολόγηση υπηρεσίας. 2.Το κοινό βασικό μοντέλο δεδομένων για την υπηρεσία (Unified Service Model). Αποτελεί τον κοινό τόπο μεταξύ ενός πλήθους διαφορετικών αλλά και συχνά επικαλυπτόμενων μοντέλων τα οποία επιχειρούν την αναπαράσταση των υπηρεσιών από τη οπτική του παρόχου. 3.Το πελατοκεντρικό μοντέλο υπηρεσιών (Customer Service Model). Αποτελεί μια επέκταση του κοινού βασικού μοντέλου, προκειμένου να συμπεριλάβει την μοντελοποίηση της οπτικής του πελάτη για την υπηρεσία. Το μοντέλο εισάγει έννοιες όπως η θυσία, η προσδοκία του πελάτη, η ανατροφοδότηση και η αντιλαμβανόμενη αξία. Οι έννοιες του μοντέλου μελετώνται και οργανώνονται βάσει των οπτικών του πλαισίου Zachman. 4.Το παράδειγμα των κοινωνικών περιγραφών υπηρεσιών (social descriptions of services). Μια νέα έκφανση για την περιγραφή υπηρεσιών προερχόμενη από τον πελάτη. Οι περιγραφές αυτές εκφράζουν τις προσδοκίες και την ανατροφοδότηση των πελατών, και ενσωματώνουν πληροφορίες από την αξιολόγησή των υπηρεσιών από τους πελάτες. Οι κοινωνικές περιγραφές των υπηρεσιών συμπληρώνουν τις περιγραφές που δημιουργούνται και διατηρούνται από τους παρόχους. Οι βασικές θεωρητικές συμβολές της εργασίας συνοψίζονται στα εξής: •Συμβάλλει στην μοντελοποίηση της υπηρεσίας, η οποία εξακολουθεί να αποτελεί ένα ενεργό ερευνητικό πεδίο και ένα απαιτητικό πεδίο εφαρμογής, μέσω της ανάπτυξης ενός ολιστικού, πελατοκεντρικού μοντέλου, καλύπτοντας την μοντελοποίηση τόσο της ίδιας της υπηρεσίας όσο και του κύκλου ζωής της. •Συμβάλλει στην ικανοποίηση της απαίτησης για πελατοκεντρικές υπηρεσίες, η οποία είναι μια από τις βασικές αρχές που διέπουν το διαδίκτυο των υπηρεσιών. •Τοποθετημένη στην περιοχή της Επιστήμης των Υπηρεσιών (Service Science) συνδυάζει τη μελέτη της υπηρεσίας τόσο από την τεχνική σκοπιά (επιστήμη υπολογιστών) όσο και από την επιχειρηματική (μάρκετινγκ, μάνατζμεντ…), μεταφέροντας έτσι γνώση από το ένα πεδίο στο άλλο. •Με την εισαγωγή ενός κοινού βασικού μοντέλου για την υπηρεσία, το οποίο καλύπτει όλες τις υφιστάμενες προσπάθειες, μειώνει τα σημασιολογικά εμπόδια διαλειτουργικότητας και συμβάλλει στην σημασιολογική διασύνδεση των υπηρεσιών, καθώς και των υφιστάμενων περιγραφών τους που κατασκευάζονται χρησιμοποιώντας διαφορετικά σημασιολογικά μοντέλα.

In cloud-native applications, a large fraction of operational failures, known as outages, result in violations of Service Level Objectives (SLOs). SLOs are defined around specific measurable characteristics: availability, throughput, frequency, response time, and quality. Four metrics, latency, traffic, errors, and saturation, ensure coverage for most outages of an application. These are often called golden signals. The dynamicity and complexity of cloud-native applications complicate Site Reliability Engineers’ (SREs) efforts in problem determination, in particular in its fault localization. The fault localization is often a try-and-error process in which SREs rely on their domain knowledge and experience. It is laborious and frequently results in long Mean Time To Resolution (MTTR) for outages. This paper describes a lightweight fault localization system, that establishes causal relationships among the golden signal service errors and error logs, and further leverages PageRank centrality of the derived causal graph for generating a ranked list of faulty microservices.

Service Level Objectives (SLOs) allow defining expected performance of cloud services, such that cloud service providers know what they guarantee and service consumers know what to expect. Most approaches focus on low-level SLOs, closely related to resources, e.g., average CPU or memory usage, and are usually bound to specific elasticity controllers. We present SLO Script, a language and accompanying framework, motivated by real-world, industrial needs to allow service providers to define complex, high-level SLOs in an orchestrator-independent manner. The main features of SLO Script include: i) novel abstractions (StronglyTypedSLO) with type safety features, ensuring compatibility between SLOs and elasticity strategies, ii) abstractions that enable decoupling of SLOs from elasticity strategies, iii) a strongly typed metrics API, and iv) an orchestrator-independent object model that enables language extensibility. We present a case study about a real-world, cloud-native application and evaluate our language while implementing a realistic Cost Efficiency SLO.

The cloud's flexibility and promise of seamless auto-scaling notwithstanding, its ability to meet service level objectives (SLOs) typically calls for some form of control in resource usage. This seemingly traditional problem gives rise to new challenges in a cloud setting, and in particular a subtle yet significant trade-off involving load-distribution decisions (the distribution of workload across available cloud resources to optimize performance), and rate limiting (the capping of individual workloads to prevent global over-commitment). This paper investigates that trade-off through the design and implementation of a real-time messaging system motivated by Internet-of- Things (IoT) applications, and demonstrates a solution capable of realizing an effective compromise. The paper's contributions are in both explicating the source of this trade-off, and in demonstrating a possible solution.

Service Level Objectives (SLOs) guide the elasticity of cloud applications, e.g., by deciding when and how much the resources provisioned to an application should be changed. Evaluating SLOs requires metrics, which can be directly measured on the application or system, or, more elaborately, be composed from multiple low-level metrics. The implementation of such metrics and SLOs, the triggering of elasticity strategies, and allowing configurability by the user deploying an application, requires a flexible middleware. In this paper, we present a middleware that provides an orchestrator-independent SLO controller for periodically evaluating SLOs and triggering elasticity strategies, while decoupling SLOs from the elasticity strategies to increase flexibility, and provider-independent services for obtaining low-level metrics and composing them into higher-level metrics. We evaluate our middleware by implementing a motivating use case, featuring a cost efficiency SLO for an application deployed on Kubernetes.

Application workload scheduling in hybrid Cloud-Edge-IoT infrastructures has been extensively researched over the last years. The recent trend of containerizing application workloads, both in the cloud and on the edge, has further fueled the need for more advanced scheduling solutions in these hybrid infrastructures. Unfortunately, most of the current approaches are not fully sensitive to the edge properties and also lack adequate support for Service Level Objective (SLO) awareness. Previously, we introduced software defined gateways (SDGs), which enable managing novel edge resources at scale. At the same time Kubernetes was initially released. In spite of not being specifically developed for the edge, Kubernetes implements many of the design principles introduced by our SDGs, making it suitable for building SDG extensions on top of it. In this paper we present Polaris Scheduler - a novel scheduling framework, which enables edge sensitive and SLO aware scheduling in the Cloud-Edge-IoT Continuum. Polaris Scheduler is being developed as a part of Linux Foundation's Centaurus project. We discuss the main research challenges, the approach, and the vision of SLO aware edge sensitive scheduling.

The rapid growth of cloud computing in cloud datacenters in recent decades greatly increases the brown energy consumption in datacenters, and hence significant increase of carbon emission that negatively impacts on the environment as well as the monetary cost. More and more cloud service providers are adopting renewable energy as the energy supply to offset the consumption of brown energy. Meanwhile, an increasing number of renewable energy generators have been built to meet the needs. However, the instability of the renewable energy cannot guarantee the support to the datacenter and the energy competition of different datacenters may lead to datacenter energy outage. In this paper, we focus on the problem of how to match different renewable energy generators to the datacenters from different cloud providers to minimize the carbon emission, monetary cost, and service level objective (SLO) violation due to renewable energy shortage. The challenges here are that the datacenters may compete in energy requesting, the renewable energy generation is not stable and the decision should be made quickly. There have been no previous efforts devoting to this problem. To solve the problem, we first test several machine learning techniques on long-term prediction accuracy on renewable energy generation and energy demand using real traces and identify SARIMA for the prediction. We then propose a multi-agent reinforcement learning based method (MARL) for each datacenter to determine how much renewable energy to request from each generator based on the predicted results. We also propose a deadline guaranteed job postponement method (DGJP) to postpone executing unurgent jobs upon insufficient renewable energy supply. The trace-driven experiments show that MARL outperforms other methods by increasing up to 35% SLO satisfaction ratio, and reducing up to 19% (0.33 billion dollars in 90 days) total monetary cost and 33% total carbon emission, and DGJP further improves the performance.

Several of today's cloud applications are spread over heterogeneous connected computing resources and are highly dynamic in their structure and resource requirements. However, serverless computing and Function-as-a-Service (FaaS) platforms are limited to homogeneous clusters and homogeneous functions. We introduce an extension of FaaS to heterogeneous computing and to support heterogeneous functions through a network of distributed heterogeneous target platforms called Function Delivery Network (FDN). A target platform is a combination of a cluster of a homogeneous computing system and a FaaS platform on top of it. FDN provides Function-Delivery-as-a-Service (FDaaS), delivering the function invocations to the right target platform. We showcase the opportunities such as collaborative execution between multiple target platforms and varied target platform's characteristics that the FDN offers in fulfilling two objectives: Service Level Objective (SLO) requirements and energy efficiency when scheduling functions invocations by evaluating over five distributed target platforms.

The advent of fast, byte-addressable persistent memory (PMEM) has fueled a renaissance in re-evaluating storage system design. Unfortunately, prior work has been unable to provide both consistent and fast performance because they rely on traditional cached or uncached approaches to system design, compromising at least one of the requirements. This paper presents DStore, a fast, tailless, and quiescent-free object store for non-volatile memory. To fulfill all three requirements, we propose a novel two-level approach, called DIPPER, which fully decouples the volatile frontend and persistent backend by leveraging the byte addressability and performance of PMEM. The novelty of our approach is in allowing the frontend and backend to operate independently and in parallel without affecting crash consistency. This not only avoids the need to quiesce the system but also allows for increased concurrency in the frontend through the use of observational equivalency. Using this approach, DStore achieves optimal scalability and low latency without compromising on crash consistency. Evaluation on Intel's Optane DC Persistent Memory Module (DCPMM) demonstrates that DStore can simultaneously provide fast performance, uninterrupted service, and low tail latency. Moreover, DStore can deliver up to 6x lower tail latency service level objectives (SLO) and up to 5x higher throughput SLO compared to state-of-the-art PMEM optimized systems.

Containers such as Docker provide a lightweight virtualization technology. They have gained popularity in developing, deploying and managing applications in and across Cloud platforms. Container management and orchestration platforms such as Kubernetes run application containers in virtual clusters that abstract the overheads in managing the underlying infrastructures to simplify the deployment of container solutions. These platforms are well suited for modern web applications that can give rise to geographic fluctuations in use based on the location of users. Such fluctuations often require dynamic global deployment solutions. A key issue is to decide how to adapt the number and placement of clusters to maintain performance, whilst incurring minimum operating and adaptation costs. Manual decisions are naive and can give rise to: over-provisioning and hence cost issues; improper placement and performance issues, and/or unnecessary relocations resulting in adaptation issues. Elastic deployment solutions are essential to support automated and intelligent adaptation of container clusters in geographically distributed Clouds. In this article, we propose an approach that continuously makes elastic deployment plans aimed at optimizing cost and performance, even during adaptation processes, to meet service level objectives (SLOs) at lower costs. Meta-heuristics are used for cluster placement and adjustment. We conduct experiments on the Australia-wide National eResearch Collaboration Tools and Resources Research Cloud using Docker and Kubernetes. Results show that with only a 0.5 ms sacrifice in SLO for the 95th percentile of response times we are able to achieve up to 44.44% improvement (reduction) in cost compared to a naive over-provisioning deployment approach.

Intelligent task placement and management of tasks in large-scale fog platforms is challenging due to the highly volatile nature of modern workload applications and sensitive user requirements of low energy consumption and response time. Container orchestration platforms have emerged to alleviate this problem with prior art either using heuristics to quickly reach scheduling decisions or AI driven methods like reinforcement learning and evolutionary approaches to adapt to dynamic scenarios. The former often fail to quickly adapt in highly dynamic environments, whereas the latter have run-times that are slow enough to negatively impact response time. Therefore, there is a need for scheduling policies that are both reactive to work efficiently in volatile environments and have low scheduling overheads. To achieve this, we propose a Gradient Based Optimization Strategy using Back-propagation of gradients with respect to Input (GOBI). Further, we leverage the accuracy of predictive digital-twin models and simulation capabilities by developing a Coupled Simulation and Container Orchestration Framework (COSCO). Using this, we create a hybrid simulation driven decision approach, GOBI*, to optimize Quality of Service (QoS) parameters. Co-simulation and the back-propagation approaches allow these methods to adapt quickly in volatile environments. Experiments conducted using real-world data on fog applications using the GOBI and GOBI* methods, show a significant improvement in terms of energy consumption, response time, Service Level Objective and scheduling time by up to 15, 40, 4, and 82 percent respectively when compared to the state-of-the-art algorithms., Comment: Accepted in IEEE Transactions on Parallel and Distributed Systems, 2021

Despite the emergence of edge computing as a key technology paradigm, there is a general lack of auto-scaling techniques specifically designed for edge computing applications. Further, existing auto-scaling solutions tailor-made for the cloud cannot be readily applied to an application running on an edge cluster. In this paper we present AMAS - a novel auto-scaling algorithm, designed specifically for an edge cluster, which allows edge devices to be automatically and seamlessly added or deleted from an edge cluster according to dynamic variations in the workload. By design, AMAS can help users and enterprises minimize the cost of infrastructure while maintaining necessary SLO (i.e., Service Level Objective) deadlines for different edge applications. We demonstrate that AMAS outperforms the state-of-the-art auto-scalers in failure-prone conditions.

Today"s world economy is becoming more of service oriented. This situation can be attributed to the increased importance and significance being witnessed in the service industry by both the developed and the developing economies of the world. In fact, growth in the service sector is often used as an indicator to measure a country"s economic growth. The service industry has since become the main stay of the economy of many nations. The service industry offers services to customers. These customers often expect value for their money. Customer satisfaction is of equal importance in the service industry just like in the product related business. Three forces dominate the prevailing marketing environment in the service sector increasing competition from private players, changing and improving technologies and continuous shifts in the regulatory environment, which has led to the growing customer sophistication. Customers have become more and more aware of their requirements and demand higher standards of services. Their perceptions and expectations are continually evolving, making it difficult for the service providers to measure and manage services effectively. The key lies in improving the service selectively, paying attention to more critical service attributes/dimensions as a part of customer service management. It is an imperative to understand how sensitive the customers are to various service attributes or dimensions. Allocating resources in the fashion that is consistent with customer priorities can enhance the effectiveness in the service operations. In addition, customer service attribute priorities need to be fully explored in service specific contexts. This paper is an attempt to explore relative importance of service quality dimensions across a "select" service context.

An increasing attention has been given to provide service level objectives (SLOs) in stream processing applications due to the performance and energy requirements, and because of the need to impose limits in terms of resource usage while improving the system utilization. Since the current and next-generation computing systems are intrinsically offering parallel architectures, the software has to naturally exploit the architecture parallelism. Implement and meet SLOs on existing applications is not a trivial task for application programmers, since the software development process, besides the parallelism exploitation, requires the implementation of autonomic algorithms or strategies. This is a system-oriented programming approach and requires the management of multiple knobs and sensors (e.g., the number of threads to use, the clock frequency of the cores, etc.) so that the system can self-adapt at runtime. In this work, we introduce a new and simpler way to define SLO in the application’s source code, by abstracting from the programmer all the details relative to self-adaptive system implementation. The application programmer specifies which parts of the code to parallelize and the related SLOs that should be enforced. To reach this goal, source-to-source code transformation rules are implemented in our compiler, which automatically generates self-adaptive strategies to enforce, at runtime, the user-expressed objectives. The experiments highlighted promising results with simpler, effective, and efficient SLO implementations for real-world applications.

Lateral transshipment and expedited shipping are the sourcing flexibility strategies that are employed to reduce inventory cost and increase service level. In this context, this study investigates the effects of lateral transshipment and expedited shipping on supply chain performance in presence of disruptions. A retail system with multiple distribution centers and multiple retailers is considered in this study. To determine the ordering policy parameters for the retailers, a simulation-based optimization approach is developed. The proposed approach considers both average supply chain cost and the worst-case supply chain cost. Moreover, decision maker’s risk attitude is modeled in the objective function as a trade-off between the average supply chain cost and the worst-case supply chain cost. In the computational experiments, two alternative system configurations, which are characterized by lateral transshipment and lateral transshipment and expediting, are compared with the base-case configuration with no sourcing flexibility. The results reveal that lateral transshipment is a cost-efficient stock-out risk mitigation strategy. Therefore, it is preferable for risk-moderate decision makers. On the other hand, for the systems with tight service level objectives, expedited shipping can be preferred since it is an effective stock-out risk mitigation strategy despite the high average supply chain cost.

QoS-aware service selection seeks to find the optimal service providers to achieve the optimization goal of a service requester, such as the maximization of utility, while satisfying global QoS requirements. Service providers are usually self-interested and have some private information, such as minimum prices, that would significantly factor into the decision making of the service requester. Thus, service requesters face a decision making dilemma with incomplete information. Recent work has used iterative combinatorial auctions to address this problem. However, such studies do not sufficiently consider that the service requester can elicit the private information from service providers by observing their actions. This can help the service selection process achieve better outcomes. In this paper, we propose a type of incentive contract that can motivate the service providers to offer the QoS and prices that the service requester prefers. Based on the incentive contracts, we propose an incentive mechanism for effective service selection. In the mechanism, a service requester offers a set of incentive contracts to the service providers and then elicits their private information based on their responses to the incentive contracts. The process is iterated until the service requester finally obtains a solution that fulfills the global QoS requirements. Experimental results show that the incentive contracts have a positive impact on both service requesters and providers and that the incentive mechanism outperforms the existing combinatorial auction-based approaches in finding optimal solutions.

As Internet applications have become more diverse in recent years, users having heavy demand for online video services are more willing to pay higher prices for better services than light users that mainly use e-mails and instant messages. This encourages the Internet Service Providers (ISPs) to explore service differentiations so as to optimize their profits and allocation of network resources. Much prior work has focused on the viability of network service differentiation by comparing with the case of a single-class service. However, the optimal service differentiation for an ISP subject to resource constraints has remained unsolved. In this work, we establish an optimal control framework to derive the analytical solution to an ISP's optimal service differentiation, i.e., the optimal service qualities and associated prices. By analyzing the structures of the solution, we reveal how an ISP should adjust the service qualities and prices in order to meet varying capacity constraints and users' characteristics. We also obtain the conditions under which ISPs have strong incentives to implement service differentiation and whether regulators should encourage such practices.

Distributed machine learning has seen immense rise in popularity in recent years. Many companies and universities are utilizing computational clusters to train and run machine learning models. Unfortunately, operating such a cluster imposes large costs. It is therefore crucial to attain as high system utilization as possible. Moreover, those who offer computational clusters as a service, apart from keeping high utilization, also have to meet the required Service Level Agreements (SLAs) for the system response time. This becomes increasingly more complex in multitenant scenarios, where the time dedicated to each task has to be limited to achieve fairness. In this work, we analyze how different parameters of the machine learning job influence the response time as well as system utilization and propose Courier. Courier is a model that, based on the type of machine learning job, can select a batch size such that the response time adheres to the Service Level Objectives (SLOs) specified, while also rendering the highest possible accuracy. We gather the data by conducting real-world experiments on a BigDL cluster. Later on, we study the influence of the factors and build several predictive models which lead us to the proposed Courier model.

In cloud computing model, resource sharing introduces major benefits for improving resource utilization and total cost of ownership, but it can create technical challenges on the running performance. In practice, orchestrators are required to allocate sufficient physical resources to each Virtual Machine (VM) to meet a set of predefined performance goals. To ensure a specific service level objective, the orchestrator needs to be equipped with a dynamic tool for assigning computing resources to each VM, based on the run-time state of the target environment. To this end, we present LOOPS, a multi-loop control approach, to allocate resources to VMs based on the service level agreement (SLA) requirements and the run-time conditions. LOOPS is mainly composed of one essential unit to monitor VMs, and three control levels to allocate resources to VMs based on requests from the essential node. A tailor-made controller is proposed with each level to regulate contention among collocated VMs, to reallocate resources if required, and to migrate VMs from one host to another. The three levels work together to meet the required SLA. The experimental results have shown that the proposed approach can meet applications' performance goals by assigning the resources required by cloud-based applications.

Scalability is promoted as a key quality feature of modern big data stream processing engines. However, even though research made huge efforts to provide precise definitions and corresponding metrics for the term scalability, experimental scalability evaluations or benchmarks of stream processing engines apply different and inconsistent metrics. With this paper, we aim to establish general metrics for scalability of stream processing engines. Derived from common definitions of scalability in cloud computing, we propose two metrics: a load capacity function and a resource demand function. Both metrics relate provisioned resources and load intensities, while requiring specific service level objectives to be fulfilled. We show how these metrics can be employed for scalability benchmarking and discuss their advantages in comparison to other metrics, used for stream processing engines and other software systems.

This study examines how human resources in the Maltese Public Service adopt new work practices in response to COVID-19 public health measures during the first wave of the pandemic. We analyze the data we collected through seven focus group discussions and ten in-depth interviews with Public Service employees and managers in a diversity of ministries and roles. Our study reveals that Public Service policies promoting remote working relied exclusively on the service's IT infrastructure. However, the ability to respond to customer needs effectively in a time of surging demand relied entirely on effective employees' access to responsive and efficient ICT support as well as employees' prior experience with remote work modes and their predisposition to change to remote working. Adopting remote working modes uncovered inherent weaknesses in the Public Service IT infrastructure that put additional strain on the Government's centralized IT support function, especially when Public Service employees adopted tools not supported by the centralized IT support. In circumstances where centralized IT support was ineffective, Public Service employees relied on their own knowledge resources which they informally shared in groups of practice or employed operant resources (or tacit knowledge) to achieve service level objectives. These observations suggest that in times when organizations respond to immediate and unprecedented change, human resources seek to adapt by relying on tacit knowledge that is shared among people in known (often informal) groups of people with a common interest or role., peer-reviewed

To solve the bottleneck problem of resource utilization and user experience quality in mobile communication networks, 5G introduces network slicing to cope with the huge resource demand of users. To further improve the quality of service for users with different needs, a new fairness definition based on service level objective is introduced. On this basis, a network slicing dynamic resource scheduling strategy based on the greedy algorithm is designed, and the actual application scenarios of slicing scheduling and user scheduling are simplified into a two-layer model, namely the slicing-user model, and combined with the greedy algorithm to make the service weight value. Combine the largest slice and the user with the highest priority, and complete the matching service. The advantage of this method is various system resources can be fairly allocated according to the same proportion to users. Through the optimal combination of each slice and user, the resources of the entire system can be fairly allocated to users with different needs. Python simulation results showed that the newly proposed network slicing dynamic resource scheduling mechanism based on the greedy algorithm can meet the different needs of users and achieve short term fairness, where the users get a fair share of the resource by each missing their SLO by a similar percentage, so as to better meet the needs of users. För att lösa flaskhalsproblemet med resursanvändning och användarupplevelsekvalitet i mobilkommunikationsnät introducerar 5G nätverksskivning för att klara användarnas enorma resursbehov. För att ytterligare förbättra servicekvaliteten för användare med olika behov införs en ny rättvisedefinition baserad på servicenivåmål. På grundval av detta utformas en dynamisk resursplaneringsstrategi för nätverksskivning baserad på den giriga algoritmen, och de faktiska applikationsscenarierna för skivningsplanering och användarschemaläggning förenklas till en tvåskiktsmodell, nämligen skivningsanvändarmodellen, och kombineras med girig algoritm för att göra tjänstens viktvärde. Kombinera den största delen och användaren med högsta prioritet och slutför motsvarande tjänst. Fördelen med denna metod är att olika systemresurser kan fördelas rättvist enligt samma andel, och genom den bästa kombinationen av varje segment och användare kan hela systemets resurser fördelas rättvist till användare med olika behov. Pythons simuleringsresultat visar att den nyligen föreslagna nätverksskärningsdynamiska resursplaneringsmekanismen baserad på den giriga algoritmen kan tillgodose användarnas olika behov och uppnå kortsiktig rättvisa där användarna får en rättvis andel av resursen genom att var och en saknar sin SLO med en liknande procentsats , för att bättre möta användarnas behov.

Serverless computing has rapidly grown following the launch of Amazon's Lambda platform. Function-as-a-Service (FaaS) a key enabler of serverless computing allows an application to be decomposed into simple, standalone functions that are executed on a FaaS platform. The FaaS platform is responsible for deploying and facilitating resources to the functions. Several of today's cloud applications spread over heterogeneous connected computing resources and are highly dynamic in their structure and resource requirements. However, FaaS platforms are limited to homogeneous clusters and homogeneous functions and do not account for the data access behavior of functions before scheduling. We introduce an extension of FaaS to heterogeneous clusters and to support heterogeneous functions through a network of distributed heterogeneous target platforms called Function Delivery Network (FDN). A target platform is a combination of a cluster of homogeneous nodes and a FaaS platform on top of it. FDN provides Function-Delivery-as-a-Service (FDaaS), delivering the function to the right target platform. We showcase the opportunities such as varied target platform's characteristics, possibility of collaborative execution between multiple target platforms, and localization of data that the FDN offers in fulfilling two objectives: Service Level Objective (SLO) requirements and energy efficiency when scheduling functions by evaluating over five distributed target platforms using the FDNInspector, a tool developed by us for benchmarking distributed target platforms. Scheduling functions on an edge target platform in our evaluation reduced the overall energy consumption by 17x without violating the SLO requirements in comparison to scheduling on a high-end target platform., Comment: Accepted at Journal of Software: Practice and Experience

Meeting tail latency Service Level Objective (SLO) as well as achieving high resource utilization is important to distributed storage systems. Recent works adopt strict priority scheduling or constant rate limiting to provide SLO guarantee but cause under-utilization resources. To address this issue, we first analyze the relationship between workload burst and latency SLO. Based on burst patterns and latency SLOs, we classify tenants into two categories: Postponement-Tolerable tenant and Postponement-Intolerable tenant. We then explore the opportunity to improve resource utilization by carefully allocating resources to each tenant type. We design Rate-Limiting-Priority scheduling algorithm to limit the impact of high priority tenants on low priority ones. Meanwhile, we propose Postponement-Aware scheduling algorithm which allows Postponement-Intolerable tenants to preempt system capacity from Postponement-Tolerable tenants. This helps to increase resource utilization. We propose a latency SLO guarantee framework Gecko. Gecko guarantees multi-tenant latency SLOs via combining the two proposed scheduling algorithms together with an admission control strategy. We evaluate Gecko with real production traces and the results show that Gecko admits 44% more tenants on average than state-of-the-art techniques meanwhile guaranteeing latency SLO.

Streaming dataflows are usually deployed as longterm services and suffer from the fluctuation in arrival rates. Many scaling controllers provide the elasticity to scale up or down the dataflow to meet the target throughput. However, for most controllers with specific target Service Level Objective (SLO), it is difficult to use a single decision to accurately complete a scaling action. Reaching the target iteratively will increase the completion time of the scaling, which leads to the performance degeneration or the wasting of resources. In this paper, we present Scaleplus, which provides pluggable services for helping iterative scaling controllers make more accurate decisions to complete the scaling rapidly. Scaleplus builds decision models incrementally without the off-line sampling procedure, which can be used out of the box, then a recommendation strategy is leveraged to recommend accurate decisions. Besides, Scaleplus can be flexibly integrated with only 3 HTTP APIs. We evaluate Scaleplus with 3 different scaling controllers on Apache Flink and Apache Heron. For the simulated 9-day traces of Twitter, Scaleplus reduces the scaling duration by 36.5%, 51.1% and 54% respectively compared with DS2, mRB and Dhalion.