A multi-resource load balancing algorithm for cloud cache systems

With the advent of cloud computing model, distributed caches have become the cornerstone for building scalable applications. Popular systems like Facebook [1] or Twitter use Memcached [5], a highly scalable distributed object cache, to speed up applications by avoiding database accesses. Distributed object caches assign objects to cache instances based on a hashing function, and objects are not moved from a cache instance to another unless more instances are added to the cache and objects are redistributed. This may lead to situations where some cache instances are overloaded when some of the objects they store are frequently accessed, while other cache instances are less frequently used. In this paper we propose a multi-resource load balancing algorithm for distributed cache systems. The algorithm aims at balancing both CPU and Memory resources among cache instances by redistributing stored data. Considering the possible conflict of balancing multiple resources at the same time, we give CPU and Memory resources weighted priorities based on the runtime load distributions. A scarcer resource is given a higher weight than a less scarce resource when load balancing. The system imbalance degree is evaluated based on monitoring information, and the utility load of a node, a unit for resource consumption. Besides, since continuous rebalance of the system may affect the QoS of applications utilizing the cache system, our data selection policy ensures that each data migration minimizes the system imbalance degree and hence, the total reconfiguration cost can be minimized. An extensive simulation is conducted to compare our policy with other policies. Our policy shows a significant improvement in time efficiency and decrease in reconfiguration cost.

StreamCloud: An Elastic Parallel-Distributed Stream Processing Engine

In recent years, applications in domains such as telecommunications, network security or large scale sensor networks showed the limits of the traditional store-then-process paradigm. In this context, Stream Processing Engines emerged as a candidate solution for all these applications demanding for high processing capacity with low processing latency guarantees. With Stream Processing Engines, data streams are not persisted but rather processed on the fly, producing results continuously. Current Stream Processing Engines, either centralized or distributed, do not scale with the input load due to single-node bottlenecks. Moreover, they are based on static configurations that lead to either under or over-provisioning. This Ph.D. thesis discusses StreamCloud, an elastic paralleldistributed stream processing engine that enables for processing of large data stream volumes. Stream- Cloud minimizes the distribution and parallelization overhead introducing novel techniques that split queries into parallel subqueries and allocate them to independent sets of nodes. Moreover, Stream- Cloud elastic and dynamic load balancing protocols enable for effective adjustment of resources depending on the incoming load. Together with the parallelization and elasticity techniques, Stream- Cloud defines a novel fault tolerance protocol that introduces minimal overhead while providing fast recovery. StreamCloud has been fully implemented and evaluated using several real word applications such as fraud detection applications or network analysis applications. The evaluation, conducted using a cluster with more than 300 cores, demonstrates the large scalability, the elasticity and fault tolerance effectiveness of StreamCloud. Resumen En los útimos años, aplicaciones en dominios tales como telecomunicaciones, seguridad de redes y redes de sensores de gran escala se han encontrado con múltiples limitaciones en el paradigma tradicional de bases de datos. En este contexto, los sistemas de procesamiento de flujos de datos han emergido como solución a estas aplicaciones que demandan una alta capacidad de procesamiento con una baja latencia. En los sistemas de procesamiento de flujos de datos, los datos no se persisten y luego se procesan, en su lugar los datos son procesados al vuelo en memoria produciendo resultados de forma continua. Los actuales sistemas de procesamiento de flujos de datos, tanto los centralizados, como los distribuidos, no escalan respecto a la carga de entrada del sistema debido a un cuello de botella producido por la concentración de flujos de datos completos en nodos individuales. Por otra parte, éstos están basados en configuraciones estáticas lo que conducen a un sobre o bajo aprovisionamiento. Esta tesis doctoral presenta StreamCloud, un sistema elástico paralelo-distribuido para el procesamiento de flujos de datos que es capaz de procesar grandes volúmenes de datos. StreamCloud minimiza el coste de distribución y paralelización por medio de una técnica novedosa la cual particiona las queries en subqueries paralelas repartiéndolas en subconjuntos de nodos independientes. Ademas, Stream- Cloud posee protocolos de elasticidad y equilibrado de carga que permiten una optimización de los recursos dependiendo de la carga del sistema. Unidos a los protocolos de paralelización y elasticidad, StreamCloud define un protocolo de tolerancia a fallos que introduce un coste mínimo mientras que proporciona una rápida recuperación. StreamCloud ha sido implementado y evaluado mediante varias aplicaciones del mundo real tales como aplicaciones de detección de fraude o aplicaciones de análisis del tráfico de red. La evaluación ha sido realizada en un cluster con más de 300 núcleos, demostrando la alta escalabilidad y la efectividad tanto de la elasticidad, como de la tolerancia a fallos de StreamCloud.

Relating goal scheduling, precedence, and memory management in and-parallel execution of logic programs

The interactions among three important issues involved in the implementation of logic programs in parallel (goal scheduling, precedence, and memory management) are discussed. A simplified, parallel memory management model and an efficient, load-balancing goal scheduling strategy are presented. It is shown how, for systems which support "don't know" non-determinism, special care has to be taken during goal scheduling if the space recovery characteristics of sequential systems are to be preserved. A solution based on selecting only "newer" goals for execution is described, and an algorithm is proposed for efficiently maintaining and determining precedence relationships and variable ages across parallel goals. It is argued that the proposed schemes and algorithms make it possible to extend the storage performance of sequential systems to parallel execution without the considerable overhead previously associated with it. The results are applicable to a wide class of parallel and coroutining systems, and they represent an efficient alternative to "all heap" or "spaghetti stack" allocation models.

OSIA: Out-of-order scheduling for in-order arriving in concurrent multi-path transfer.

One major problem of concurrent multi-path transfer (CMT) scheme in multi-homed mobile networks is that the utilization of different paths with diverse delays may cause packet reordering among packets of the same ?ow. In the case of TCP-like, the reordering exacerbates the problem by bringing more timeouts and unnecessary retransmissions, which eventually degrades the throughput of connections considerably. To address this issue, we ?rst propose an Out-of-order Scheduling for In-order Arriving (OSIA), which exploits the sending time discrepancy to preserve the in-order packet arrival. Then, we formulate the optimal traf?c scheduling as a constrained optimization problem and derive its closedform solution by our proposed progressive water-?lling solution. We also present an implementation to enforce the optimal scheduling scheme using cascaded leaky buckets with multiple faucets, which provides simple guidelines on maximizing the utilization of aggregate bandwidth while decreasing the probability of triggering 3 dupACKs. Compared with previous work, the proposed scheme has lower computation complexity and can also provide the possibility for dynamic network adaptability and ?ner-grain load balancing. Simulation results show that our scheme signi?cantly alleviates reordering and enhances transmission performance.

StreamCloud: An elastic and scalable data streaming system

Many applications in several domains such as telecommunications, network security, large scale sensor networks, require online processing of continuous data lows. They produce very high loads that requires aggregating the processing capacity of many nodes. Current Stream Processing Engines do not scale with the input load due to single-node bottlenecks. Additionally, they are based on static con?gurations that lead to either under or over-provisioning. In this paper, we present StreamCloud, a scalable and elastic stream processing engine for processing large data stream volumes. StreamCloud uses a novel parallelization technique that splits queries into subqueries that are allocated to independent sets of nodes in a way that minimizes the distribution overhead. Its elastic protocols exhibit low intrusiveness, enabling effective adjustment of resources to the incoming load. Elasticity is combined with dynamic load balancing to minimize the computational resources used. The paper presents the system design, implementation and a thorough evaluation of the scalability and elasticity of the fully implemented system.

Evaluación de la tecnología CAES como sistema de almacenamiento de energía eólica. Caso de estudio en España

Se presenta a continuación un modelo de una planta del almacenamiento de energía mediante aire comprimido siguiendo un proceso adiabático. En esta planta la energía eólica sobrante se usa para comprimir aire mediante un tren de compresión de 25 MW, el aire comprimido será después almacenado en una caverna de sal a 770 metros de profundidad. La compresión se llevará a cabo por la noche, durante 6 horas, debido a los bajos precios de electricidad. Cuando los precios de la electricidad suben durante el día, el aire comprimido es extraído de la caverna de sal y es utilizado para producir energía en un tren de expansión de 70 MW durante 3 horas. La localización elegida para la planta es el norte de Burgos (Castilla y León, España), debido a la coincidencia de la existencia de muchos parques eólicos y una formación con las propiedades necesarias para el almacenamiento. El aspecto más importante de este proyecto es la utilización de un almacenamiento térmico que permitirá aprovechar el calor de la compresión para calentar el aire a la entrada de la expansión, eliminando combustibles fósiles del sistema. Por consiguiente, este proyecto es una atractiva solución en un posible futuro con emisiones de carbono restringidas, cuando la integración de energía renovable en la red eléctrica supone un reto importante. ABSTRACT: A model of an adiabatic compressed air energy storage plant is presented. In this plant surplus wind energy is used to compress air by means of a 25 MW compression train, the compressed air will be later stored in a salt cavern at 770 meters depth. Compression is carried out at night time, during 6 hours, because power prices are lower. When power prices go up during the day, the compressed air is withdrawn from the salt cavern and is used to produce energy in an expansion train of 70 MW during 3 hours. The chosen location for the plant is in the north of Burgos (Castilla y León, Spain), due to both the existence of several wind farms and a suitable storage facility with good properties at the same place. The relevance of this project is that it is provided with a thermal storage, which allows using the generated heat in the compression for re-heating the air before the expansion, eliminating fossil fuels from the system. Hence, this system is an attractive load balancing solution in a possibly carbon-constrained future, where the integration of renewable energy sources into the electric grid is a major challenge.

Aplicación de la metodología GloBeM al servicio de cloud Hydra

El presente Proyecto de Fin de Máster consiste en crear una herramienta software capaz de monitorizar y gestionar la actividad de Hydra, una herramienta de gestión de entornos distribuidos, para que su estrategia de balanceo de carga se adecúe al modelo creado por GloBeM, una metodología de análisis de entornos distribuidos. GloBeM, que es una metodología externa, puede analizar y crear un modelo de máquina de estados finitos a partir de un sistema distribuido concreto. Hydra, una herramienta también externa, es un sistema de gestión de entornos cloud recientemente desarrollado y de código abierto, con un sistema de balanceo de carga efectivo pero algo limitado. El software construido recoge el modelo creado por GloBeM y lo analiza. A partir de ahí, monitoriza en tiempo real y a una frecuencia determinada la actividad de Hydra y el sistema cloud que ésta gestiona, y reconfigura sus parámetros para que su desempeño se ciña a lo estipulado por el modelo de GloBeM, extendiendo así el sistema de balanceo de carga original de Hydra.---ABSTRACT---This Master's Thesis Project involves creating a software able to monitor and manage the activity of Hydra, a tool for managing distributed environments, in order to adjust its load balancing strategy to the model created by GloBeM, an analysis methodology for distributed environments. GloBeM, which is an external methodology, can analyse and create a finite-state machine model from a particular cloud system. Hydra, also an external tool, is an open source management system for cloud environments recently developed, with a relatively limited system of load balancing. The created software gets the model created by GloBeM as an input and analyses it. From there, it monitors in real time and at a certain frequency Hydra’s activity and the cloud system that it manages, and reconfigures its parameters to adjust its performance to the stipulations by the GloBeM’s model, extending Hydra's original load balancing system.