Patricia Stolf

A Runtime Framework for Energy Efficient HPC Systems without a Priori Knowledge of Applications

The rising computing demands of scientific endeavors often require the creation and management of High Performance Computing (HPC) systems for running experiments and processing vast amounts of data. These HPC systems generally operate at peak performance, consuming a large quantity of electricity, even though their workload varies over time. Understanding the behavioral patterns (i.e., phases) of HPC systems during their use is key to adjust performance to resource demand and hence improve the energy efficiency. In this paper, we describe (i) a method to detect phases of an HPC system based on its workload, and (ii) a partial phase recognition technique that works cooperatively with on-the-fly dynamic management. We implement a prototype that guides the use of energy saving capabilities to demonstrate the benefits of our approach. Experimental results reveal the effectiveness of the phase detection method under real-life workload and benchmarks. A comparison with baseline unmanaged execution shows that the partial phase recognition technique saves up to 15% of energy with less than 1% performance degradation.

Energy-efficient, thermal-aware modeling and simulation of data centers

This paper describes the CoolEmAll project and its approach for modeling and simulating energy-efficient and thermal-aware data centers. The aim of the project was to address energy-thermal efficiency of data centers by combining the optimization of IT, cooling and workload management. This paper provides a complete data center model considering the workload profiles, the applications profiling, the power model and a cooling model. Different energy efficiency metrics are proposed and various resource management and scheduling policies are presented. The proposed strategies are validated through simulation at different levels of a data center.

DNA-inspired scheme for building the energy profile of HPC systems

Energy usage is becoming a challenge for the design of next generation large scale distributed systems. This paper explores an innovative approach of profiling such systems. It proposes a DNA-like solution without making any assumptions on the running applications and used hardware. This profiling based on internal counters usage and energy monitoring allows to isolate specific phases during the execution and enables some energy consumption control and energy usage prediction. First experimental validations of the system modeling are presented and analyzed.

Beyond CPU Frequency Scaling for a Fine-grained Energy Control of HPC Systems

Modern high performance computing subsystems (HPC) - including processor, network, memory, and IO - are provided with power management mechanisms. These include dynamic speed scaling and dynamic resource sleeping. Understanding the behavioral patterns of high performance computing systems at runtime can lead to a multitude of optimization opportunities including controlling and limiting their energy usage. In this paper, we present a general purpose methodology for optimizing energy performance of HPC systems considering processor, disk and network. We rely on the concept of execution vector along with a partial phase recognition technique for on-the-fly dynamic management without any a priori knowledge of the workload. We demonstrate the effectiveness of our management policy under two real-life workloads. Experimental results show that our management policy in comparison with baseline unmanaged execution saves up to 24% of energy with less than 4% performance overhead for our real-life workloads.

Energy-efficient and thermal-aware resource management for heterogeneous datacenters

We propose in this paper to study the energy-, thermal- and performance-aware resource management in heterogeneous datacenters. Witnessing the continuous development of heterogeneity in datacenters, we are confronted with their different behaviors in terms of performance, power consumption and thermal dissipation: indeed, heterogeneity at server level lies both in the computing infrastructure (computing power, electrical power consumption) and in the heat removal systems (different enclosure, fans, thermal sinks). Also the physical locations of the servers become important with heterogeneity since some servers can (over)heat others. While many studies address independently these parameters (most of the time performance and power or energy), we show in this paper the necessity to tackle all these aspects for an optimal resource management of the computing resources. This leads to improved energy usage in a heterogeneous datacenter including the cooling of the computer rooms. We build our approach on the concept of heat distribution matrix to handle the mutual influence of the servers, in heterogeneous environments, which is novel in this context. We propose a heuristic to solve the server placement problem and we design a generic greedy framework for the online scheduling problem. We derive several single-objective heuristics (for performance, energy, cooling) and a novel fuzzy-based priority mechanism to handle their tradeoffs. Finally, we show results using extensive simulations fed with actual measurements on heterogeneous servers.

Multi-objective Scheduling for Heterogeneous Server Systems with Machine Placement

Heterogeneous servers are becoming prevalent in many high-performance computing environments, including clusters and data enters. In this paper, we consider multi-objective scheduling for heterogeneous server systems to optimize simultaneously the application performance, energy consumption and thermal imbalance. First, a greedy online framework is presented to allow the scheduling decisions to be made based on any well-defined cost function. To tackle the possibly conflicting objectives, we propose a fuzzy-based priority approach for exploring the tradeoffs of two or more objectives at the same time. Moreover, we present a heuristic algorithm for the static placement of physical machines in order to reduce the maximum temperature at the server outlets. Extensive simulations based on an emerging class of high-density server system have demonstrated the effectiveness of our proposed approach and heuristics in optimizing multiple objectives while achieving better thermal balance.

Component-Based Autonomic Management for Legacy Software

Distributed software environments are increasingly complex anddifficult to manage, as they integrate various legacy software withspecific management interfaces. Moreover, the fact that management tasks are performed by humans leads to many configuration errors and low reactivity. This is particularly true in medium or large-scale distributed infrastructures. To address this issue, we explore the design and implementation of an autonomic management system. The main principle is to wrap legacy software pieces in components in order to administrate a software infrastructure as a component architecture. In order to help the administrators defining autonomic management policies, we introduce high-level formalisms for the specification of deployment and management policies. This chapter describes the design and implementation of such a system and its evaluation with different use cases.

Non-Intrusive Autonomic Approach with Self-Management Policies Applied to Legacy Infrastructures for Performance Improvements

The growing complexity of large IT facilities involves important time and effort costs to operate and maintain. Autonomic computing gives a new approach in designing distributed architectures that manage themselves in accordance with high-level objectives. The main issue is that existing architectures do not necessarily follow this new approach. The motivation is to implement a system that can interface heterogeneous components and platforms supplied by different vendors in a non-intrusive and generic manner. The goal is to increase the intelligence of the system by actively monitoring its state and autonomously taking corrective actions without the need to modify the managed system. In this paper, the authors focus on modeling software and hardware architectures as well as describing administration policies using a graphical language inspired from UML. The paper demonstrates that this language is powerful enough to describe complex scenarios and evaluates some self-management policies for performance improvement on a distributed computational jobs load balancer over a grid.

Autonomic Management for Grid Applications

Distributed software environments are increasingly complex and difficult to manage, as they integrate various legacy software with specific management interfaces. Moreover, the fact that management tasks are performed by humans leads to many configuration errors and low reactivity. This is particularly true in medium or large-scale grid infrastructures. To address this issue, we developed Jade, a middleware for self-management of distributed software environments. In this paper, we report on our experiments in using Jade for the management of grid applications.

An Energy Efficient Approach to Virtual Machines Management in Cloud Computing

Cloud computing platforms are well established as the reference infrastructure for flexibility allowing business to scale according to user demands. However, several challenges remain to be addressed. Scaling up the infrastructure to match up the resource demand of the users is relatively straight forward. However in order to save up on power consumption, cloud providers have to dynamically consolidate the virtual machines (VM), so that only a reduced subset of hosts remains powered on. To solve this issue, the cloud scheduler, which handles the VM allocations, has to balance between providing enough resource to each user, and reducing the overall power consumption of the cloud. In this paper we investigate the VM allocation and reallocation inside a cloud in order to save energy while maintaining the resources required by users. Such actions have to be made in order to minimize the number of hosts powered on, while limiting concurrent migrations of virtual machines, and in a reasonable computational time. We propose to effectively consolidate virtual machines with an approach handling the reallocation, migration and host management problems. Our approach has been implemented in OpenNebula, and experimentally compared with its default approach.