Leveraging location information in location-based services leads to improving service utility through geocontextualization. However, this raises privacy concerns as new knowledge can be inferred from location records, such as user's home and work places, or personal habits. Although Location Privacy Protection Mechanisms (LPPMs) provide a means to tackle this problem, they often require manual configuration posing significant challenges to service providers and users. Moreover, their impact on data privacy and utility is seldom assessed. In this paper, we present PULP, a model-driven system which automatically provides user-specific privacy protection and contributes to service utility via choosing adequate LPPM and configuring it. At the heart of PULP is nonlinear models that can capture the complex dependency of data privacy and utility for each individual user under given LPPM considered, i.e., Geo-Indistinguishability and Promesse. According to users' preferences on privacy and utility, PULP efficiently recommends suitable LPPM and corresponding configuration. We evaluate the accuracy of PULP's models and its effectiveness to achieve the privacy-utility trade-off per user, using four real-world mobility traces of 770 users in total. Our extensive experimentation shows that PULP ensures the contribution to location service while adhering to privacy constraints for a great percentage of users, and is orders of magnitude faster than non-model based alternatives.

@en

To ensure the scalability of big data analytics, approximate MapReduce platforms emerge to explicitly trade off accuracy for latency. A key step to determine optimal approximation levels is to capture the latency of big data jobs, which is long deemed challenging due to the complex dependency among data inputs and map/reduce tasks. In this paper, we use matrix analytic methods to derive stochastic models that can predict a wide spectrum of latency metrics, e.g., average, tails, and distributions, for approximate MapReduce jobs that are subject to strategies of input sampling and task dropping. In addition to capturing the dependency among waves of map/reduce tasks, our models incorporate two job scheduling policies, namely, exclusive and overlapping, and two task dropping strategies, namely, early and straggler, enabling us to realistically evaluate the potential performance gains of approximate computing. Our numerical analysis shows that the proposed models can guide big data platforms to determine the optimal approximation strategies and degrees of approximation.

@en

Replicating redundant requests has been shown to be an effective mechanism to defend application performance from high capacity variability - the common pitfall in the cloud. While the prior art centers on single-tier systems, it still remains an open question how to design replication strategies for distributed multi-tier systems, where interference from neighboring workloads is entangled with complex tier interdependency. In this paper, we design a first of its kind PArtial REplication system, sPARE, that replicates and dispatches read-only workloads for multi-tier web applications, determining replication factors per tier. The two key components of sPARE are (i) the variability-aware replicator that coordinates the replication levels on all tiers via an iterative searching algorithm, and (ii) the replication-aware arbiter that uses a novel token-based arbitration algorithm (TAD) to dispatch requests in each tier. We evaluate sPARE on web serving and web searching applications, i.e., MediaWiki and Solr, deployed on our private cloud testbed. Our results based on various interference patterns and traffic loads show that sPARE is able to improve the tail latency of MediaWiki and Solr by a factor of almost 2.7x and 2.9x, respectively.

@en

Offering consistent low latency remains a key challenge for distributed applications, especially when deployed on the cloud where virtual machines (VMs) suffer from capacity variability caused by colocated tenants. Replicating redundant requests were shown to be an effective mechanism to defend application performance from high capacity variability. While the prior art centers on single-tier systems, it still remains an open question how to design replication strategies for distributed multi-tier systems. In this paper, we design a first of its kind PArtial REplication system, sPARE, that replicates and dispatches read-only workloads for distributed multi-tier web applications The two key components of sPARE are (i) the variability-aware replicator that coordinates the replication levels on all tiers via an iterative searching algorithm, and (ii) the replication-aware arbiter that uses a novel token-based arbitration algorithm (TAD) to dispatch requests in each tier. We evaluate sPARE on web serving and web searching applications, i.e., MediaWiki and Solr, the former deployed on our private cloud and the latter in the wild on Amazon EC2. Our results based on various interference patterns and traffic loads show that sPARE is able to improve the tail latency of MediaWiki and Solr by a factor of almost <formula> <tex>$2.7x$</tex> </formula> and <formula> <tex>$2.9x$</tex> </formula>, respectively.

@en

Nowadays, numerous programming languages and frameworks offer concurrency based on the actor model. Among the actor libraries for the Java Virtual Machine, Akka is the most used one, as it is employed in various parallel and distributed applications and frameworks. Unfortunately, despite the spread of actors libraries, Akka in particular, existing profiling tools are not very effective at identifying performance drawbacks in applications using actors. In this tool paper, we aim at filling this gap by presenting AkkaProf, a profiling tool for Akka actors. AkkaProf provides detailed metrics on actor utilization and on the communication between them, two fundamental aspects of actor-based applications that are overlooked by other profilers. AkkaProf aids performance analysis in several applications and frameworks in both parallel and distributed environments.

@en

Concurrent execution of multiple applications leads to varying partial utilization of shared resources. Understanding system behavior in these conditions is essential for making concurrent execution efficient. Unfortunately, anticipating behavior of shared resources at partial utilization in complex systems is difficult, realistic experiments that reproduce and examine such behavior are therefore needed. To facilitate experiments at partial utilization, we present a tool that accurately controls the processor utilization of arbitrary concurrent workloads, either establishing constant partial load or replaying a variable load trace. We validate the ability of the tool to enforce the configured partial utilization on multiple platforms, and use the tool to collect novel information on system behavior at partial utilization levels. In detail, our experiments show how to examine the complex relationship between utilization and throughput, useful for tasks such as performance debugging or system dimensioning, and we show this relationship for the DaCapo benchmarks. Further, we show that CPU pinning (a technique used to improve workload isolation) can benefit from dynamic response to system utilization, improving system efficiency with partial utilization. Finally, we show that the overhead of virtualization also changes with partial utilization and CPU allocation.@en

Energy consumption in data centres accounts for a significant proportion of national energy usage in many countries. One approach for reducing energy consumption is to improve the server usage efficiency via workload consolidation. However, there are two primary reasons why this is not done to a large extent. The first reason is that greater consolidation could result in violations of Service Level Agreements (SLAs) if resources are over-utilised. The second reason is that users specify the requirements of a virtual machine (VM) based on the maximum estimated usage for each resource over the whole life span of the VM, and usually over-estimate these maximum values to avoid possible contract violations. Typically, the VM will have significantly lower resource usage in most time periods. Recently, a number of methods have been proposed to predict resource usage of VMs. We show that although these prediction techniques are efficient when their performances are measured using well known metrics, a low prediction error can still result in significant violations of SLAs if not handled properly during workload allocation. Our results emphasise the importance of analysing workload prediction in conjunction with workload allocation techniques. We examine the impact of using predicted resource usage for optimal server consolidation. We investigate the occurrences of over-utilised resources on servers due to under-predicted resource usage. We propose methods to reduce the likelihood of such occurrences, both through the enforcement of safety capacities on the server side, and through biasing towards over-prediction on the VM side. The results indicate that an appropriate balance can be found between energy savings and non-violation of SLAs.

@en

Congestion control becomes indispensable in highly utilized consolidated networks running demanding applications. In this paper, proactive congestion management schemes for Clos networks are described and evaluated. The key idea is to move the congestion avoidance burden from the data fabric to a scheduling network, which isolates flows using per-flow request counters. The scheduling network comprises per-output arbiters that grant data packets after reserving space for them in the buffer memories in front of fabric outputs. Computer simulations show that this strategy eliminates head-of-line (HOL) blocking and its adversarial effects throughout the fabric, without having to drop packets. In particular, a simplified model describes this result as a synergy between proactive buffer reservations and fine-grained multipath routing. Two alternative designs are presented. The first one places all arbiters in a central control unit, is simpler, and has superior performance. The second is more scalable by distributing the arbiters over the switching elements of the Clos network and by routing the control messages to and from endpoint adapters via multiple paths. Computer simulations of the complete system demonstrate high throughput and low latency under any number of congested outputs. Weighted max-min fair allocation of fabric-output link bandwidth is also demonstrated. Furthermore, delay breakdowns show that the time that packets wait in fabric and resequencing buffers is minimized as a result of the reduced (and equalized across all fabric paths) in-fabric contention. Finally, the high throughput capability of the system is corroborated by a Markov chain analysis of output buffer credits.@en

Big-data applications are becoming the core of today's business operations, featuring complex data structures and high task fan-out. According to the publicly available Google trace, more than 40% of big-data jobs do not reach successful completion. Interestingly, a significant portion of tasks of such failed jobs undergo multiple types of repetitive failed executions and consume a non-negligible amount of resources. To conserve resources for big-data clusters, it is imperative to capture such failed tasks of failed jobs, a very challenging problem due to multiple types of failures associated with tasks and highly uneven tasks distribution. In this paper, we develop an on-line two-level Neural Network (NN) model which can accurately untangle the complex dependencies among tasks and jobs, and predict their execution classes in an extremely dynamic and heterogeneous system. Our proposed NN model predicts first the job class, and secondly three classes of failed tasks of failed jobs, based on a sliding learning window. Furthermore, we develop resource conservation policies that terminate failed tasks of failed jobs after a grace period that is derived from prediction confidences and task execution times. Overall, evaluating our results on a Google cluster trace, we are able to accurately capture failures of failures at big-data clusters, mitigate false negative tasks to 1%, and efficiently save system resources, achieving significant reductions of CPU, memory and disk consumption - as high as 49%.

@en

Visualization technology powers up the cloud computing paradigm and inevitably raises concerns about performance isolation of collocated virtual machines (VM). It is imperative for public cloud providers to guarantee performance targets for tenants' VMs while respecting strict business confidentiality, e.g., having no information on applications nor their performance. A large body of related work addresses the challenges of detecting performance interferences by leveraging client's quality of service (QoS) metrics, such as latency, and additional profiling servers. Whereas to assist cloud providers, we resort to an on-line blackbox approach based on throttling that detects a wide range of resource contentions with no cooperation need from the virtual machines. We focus on different resource metrics and actively monitor them from the hypervisor in fine time granularity at low cost. To detect resource contention, we propose a three-phase algorithm: an alarm phase, to identify statistical outliers in the victim's VM resource metrics; a passive diagnosis phase, to match the current sample to historical behaviors; and, an active learning phase, to discern contentions from application phase changes via throttling. We evaluate our algorithm on a prototype running Wikimedia as victim application across a set of VMs collocated with neighboring VMs running resource hoggers, i.e. PARSEC and Cachebench. Our extensive experimental results show that we can reach an average detection accuracy above 90% while limiting the performance degradation experienced by offender workloads to short learning phases.

@en

Real-time processing of big data is becoming one of the core operations in various areas, such as social networks and anomaly detection. Thanks to the rich information of the data, multiple queries can be executed to analyse the data and discover a variety of business values. It is very typical that a cluster infrastructure running for example a Spark Streaming data stream processing system would execute multiple queries simultaneously. To enable multiple queries being answered from the same data concurrently, it is important to effectively allocate the CPU-cores of the underlying infrastructure to the queries, meanwhile adhering to the latency constraints of the individual queries. In this paper, we consider the problem of allocating CPU-cores in a Spark Streaming infrastructure in the context of two types of queries, namely primary and optional, that are associated with high-and low-priority analysis, respectively. We develop a controller, iBLOC, that adjusts the block sizes of streaming jobs on the fly and the parallelism level of jobs, according to the input data rates and the query priorities. Our evaluation shows that we can achieve significant CPU-core savings from the primary query type such that multiple queries can run together without impairing their latency constraints, in comparison to a static block-sizing scheme.

@en

Message exchange is a central activity in distributed computing frameworks. Nevertheless, past research has paid little attention on profiling techniques and tools for endpoint communication. In this paper, we fill this gap by introducing a new fine-grained profiler for endpoints and communication between them in distributed systems. Our tool aids efficient analysis, optimization, and tuning of endpoint communication in distributed frameworks. The paper presents an overview of our profiler, discusses employed profiling techniques and collected metrics, and shows preliminary results in profiling a well-known distributed computing framework.

@en

Peak loads are a challenge for streaming systems. Here we present Dslash, a latency-driven controller that keeps and processes as much data as the system resources allow in order to meet a target latency.

@en

Virtualization has become a mainstream technology that allows efficient and safe resource sharing in data centers. In this paper, we present a large scale workload characterization study of 90K virtual machines hosted on 8K physical servers, across several geographically distributed corporate data centers of a major service provider. The study focuses on 19 days of operation and focuses on the state of the practice, i.e., how virtual machines are deployed across different physical resources with an emphasis on processors and memory, focusing on resource sharing and usage of physical resources, virtual machine life cycles, and migration patterns and their frequencies. This paper illustrates that indeed there is a huge tendency in over-provisioning CPU and memory resources while certain virtualization features (e.g., migration and collocation) are used rather conservatively, showing that there is significant room for the development of policies that aim to reduce operational costs in data centers.

@en

Workload redundancy emerges as an effective method to guarantee quality of service (QoS) targets, especially tail latency, in environments with strong capacity variability such as clouds. Nevertheless mostly single-tier replication strategies have been studied, while multi-tier architectures are a very popular choice in cloud deployed applications. In this poster we raise the awareness on the challenges of devising a multi-tier replication strategy and its possible benefits through a 2.3X tail latency improvement on an exemplary real-world application.

@en

As modern service systems are pressured to provide competitive prices via cost-effective capacity planning, especially in the paradigm of cloud computing, service level agreements (SLAs) end up becoming ever more sophisticated, i.e., fulfilling targets of different percentiles of response times. However, it is no mean feat to predict even the average response times of real systems, or even abstracted queueing systems that typically simplify system details, and it gets even more complicated when trying to manage SLAs defined by various percentiles of response times. To efficiently capture these different percentiles, we first develop a novel and autonomic methodology - termed Burst Based Simulation, which combines burst profiling on real systems with complex, state-dependent simulations. Moreover, based on our methodology, we construct an analysis on SLA management: the prediction of SLA violations given a certain request pattern. We evaluate our approach on two types of service systems, virtualized and bare-metal, with wide ranges of SLAs and traffic loads. Our evaluation results show that our methodology is able to achieve an average error below 15% when predicting different response time percentiles, and accurately capture SLA violations.

@en

To operate systems cost-effectively, cloud providers not only multiplex applications on the shared infrastructure but also dynamically allocate available resources, such as power and cores. Data intensive applications based on the MapReduce paradigm rapidly grow in popularity and importance in the Cloud. Such big data applications typically have high fan-out of components and workload dynamics. It is no mean feat to deploy and further optimize application performance within (stringent) resource budgets. In this paper, we develop a novel solution, OptiCA, that eases the deployment of big data applications on cloud and the control of application components so that desired performance metrics can be best achieved for any given resource budgets, in terms of core capacities. The control algorithm of OptiCA distributes the available core budget across co-executed applications and components, based on their 'effective' demands obtained through non-intrusive profiling. Our proposed solution is able to achieve robust performance, i.e., with very minor degradation, in cases where resource budget decreases rapidly.

@en

The energy-performance optimization of datacenters becomes ever challenging, due to heterogeneous workloads featuring different performance constraints. In addition to conventional web service, MapReduce presents another important workload class, whose performance highly depends on data availability/locality and shows different degrees of delay sensitivities, such as batch vs. interactive MapReduce. However, current energy optimization solutions are mainly designed for a subset of these workloads and their key features. Here, we present an energy minimization framework, in particular, a concave minimization problem, that specifically considers time variability, data locality, and delay sensitivity of web, batch-, and interactive-MapReduce. We aim to maximize the usage of MapReduce servers by using their spare capacity to run non-MapReduce workloads, while controlling the workload delays through the execution of MapReduce tasks, in particular batch ones. We develop an optimal algorithm with complexity O(T2) in case of perfect workload information, T being the length of the time horizon in number of control windows, and derive the structure of optimal policy for the case of uncertain workload information. Using extensive simulation results, we show that the proposed methodology can efficiently minimize the datacenter energy cost while fulfilling the delay constraints of workloads.@en

As energy costs become increasingly greater contributors to a cloud provider's overall costs, it is important for the cloud to recoup these energy costs from its tenants for profitability via appropriate pricing design. The poor predictability of real-world tenants' demand and demand responses (DRs) make such pricing design a challenging problem. We formulate a leader-follower game-based cloud pricing framework with the goal of maximizing cloud's profit. The key distinguishing aspect of our approach is our emphasis on modeling both the cloud and its tenants as working with low predictability in their inputs. Consequently, we model them as employing myopic control with short-term predictive models. Our empirical evaluation using tenant trace from IBM production data centers shows that (i) cloud's profit and VM prices are sensitive to the tradeoffs between its energy costs, tenant's demand and DR, and (ii) the cloud's estimation of tenants' demands/DR may significantly affect its profitability.

@en