Grappe de serveurs | EPFL Graph Search

On parle de grappe de serveurs, de cluster, de groupement de serveurs ou de ferme de calcul (computer cluster en anglais) pour désigner des techniques consistant à regrouper plusieurs ordinateurs indépendants appelés nœuds (node en anglais), afin de permettre une gestion globale et de dépasser les limitations d'un ordinateur pour :

augmenter la disponibilité ;
faciliter la montée en charge ;
permettre une répartition de la charge ;
faciliter la gestion des ressources (processeur, mémoire vive, disques durs, bande passante réseau).

La création de petites grappes de serveurs est un procédé peu coûteux, consistant à grouper plusieurs ordinateurs en réseau, qui apparaîtront comme un ordinateur unique doté de plus de puissance du processeur, d'espace de stockage, de mémoire vive, etc., particulièrement utilisé pour les calculs parallèles. Cet usage optimisé des ressources permet la répartition des traitements sur les différents nœuds. L'un des principaux avantages est d'éviter l

Robotic clusters: Multi-robot systems as computer clusters A topological map merging demonstration

Sarvenaz Choobdar, Ali Marjovi

In most multi-robot systems, an individual robot is not capable of solving computationally hard problems due to lack of high processing power. This paper introduces the novel concept of robotic clusters to empower these systems in their problem solving. A robotic cluster is a group of individual robots which are able to share their processing resources, therefore, the robots can solve difficult problems by using the processing units of other robots. The concept, requirements, characteristics and architecture of robotic clusters are explained and then the problem of “topological map merging” is considered as a case study to describe the details of the presented idea and to evaluate its functionality. Additionally, a new parallel algorithm for solving this problem is developed. The experimental results proved that the robotic clusters remarkably speedup computations in multi-robot systems. The proposed mechanism can be used in many other robotic applications and has the potential to increase the performance of multi-robot systems especially for solving problems that need high processing resources.

Elsevier2012

Hybrid, Job-Aware, and Preemptive Datacenter Scheduling

Pamela Isabel Delgado Borda

Scheduling in datacenters is an important, yet challenging problem. Datacenters are composed of a large number, typically tens of thousands, of commodity computers running a variety of data-parallel jobs. The role of the scheduler is to assign cluster resources to jobs, which is not trivial due to the large scale of the cluster, as well as the high scheduling load (tens of thousands of scheduling decisions per second). Additionally to scalability, modern datacenters face increasingly heterogeneous workloads composed of long batch jobs, e.g., data analytics, and latency-sensitive short jobs, e.g., operations of user-facing services. In such workloads, and especially if the cluster is highly utilized, it is challenging to avoid short running jobs getting stuck behind long running jobs, i.e. head-of-line blocking. Schedulers have evolved from being centralized (one single scheduler for the entire cluster) to distributed (many schedulers that take scheduling decisions in parallel). Although distributed schedulers can handle the large-scale nature of datacenters, they trade scheduling latency for accuracy. The complexity of scheduling in datacenters is exacerbated by the data-parallel nature of the jobs. That is, a job is composed of multiple tasks and the job completes only when all of its tasks complete. A scheduler that takes into account this fact, i.e. job-aware, could use this information to provide better scheduling decisions. Furthermore, to improve the quality of their scheduling decisions, most of datacenter schedulers use job runtime estimates. Obtaining accurate runtime estimates is, however, far from trivial, and erroneous estimates may lead to sub-optimal scheduling decisions. Considering these challenges, in this dissertation we argue the following: (i) that a hybrid centralized/distributed design can get the best of both worlds by scheduling long jobs in a centralized way and short jobs in a distributed way; (ii) such a hybrid scheduler can avoid head-of-line blocking and provide job-awareness by dynamically partitioning the cluster for short and long jobs and by executing a job to completion once it started; (iii) a scheduler can dispense with runtime estimates by sharing the resources of a node with preemption, and load balancing jobs among the nodes.

EPFL2018

An Architecture for Load Balance in Computer Cluster Applications

Laurent Bindschaedler

Amid a data revolution that is transforming industries around the globe, computing systems have undergone a paradigm shift where many applications are scaled out to run on multiple computers in a computing cluster. As the storage and processing capabilities of a single machine are unable to keep pace with the amount of data, companies turn to distributed solutions to organize, persist, and analyze this Big Data. These new software solutions divide datasets into partitions that are processed in parallel on separate machines. A common problem in current cluster computing frameworks is load imbalance and limited parallelism due to skewed data distributions, processing times, and machine speeds. Load imbalance occurs when a few machines have more work and take longer to finish while the others remain idle, resulting in reduced overall performance and low resource utilization. The underlying cause for these issues is that data locality, where machines process the data stored locally, leads to tight coupling between a partition and the machine on which it is placed. This dissertation proposes a novel scatter architecture for computer cluster applications that effectively addresses the load imbalance problem and improves resource utilization. Scatter systems abandon data locality in favor of load balance. Existing systems first target locality, bringing computation to the data, and only then attempt to minimize load imbalance. Instead, we disregard any locality concerns and focus solely on achieving load balance by scattering all data across machines and bringing data to the computation as needed. The scatter architecture disaggregates compute and storage resources and pools all resources together across machines. We ensure storage load balance by spreading the data for each partition in small blocks across all storage devices and allow machines to retrieve data using an efficient, decentralized scheme. Scatter systems achieve compute load balance at runtime by dynamically adjusting the parallelism within a partition, either through work-sharing or by offloading background tasks to other machines. We design and implement three cluster applications inspired by the scatter architecture: a graph processing system that scales out using secondary storage, a general-purpose analytics framework, and a filesystem substrate that mitigates load imbalance for existing distributed databases. We demonstrate that each application can gracefully handle significant load imbalance with minimal performance degradation and that these applications can perform up to an order of magnitude faster than existing systems.

EPFL2020