New PFC, TFG, TFM and collaboration proposals at GreenLSI

The GreenLSI team is offering some Master Thesis positions (also known as “Proyecto Final de Carrera”, “Trabajo Fin de Grado” and “Trabajo Fin de Máster”) and collaboration oportunities in the area of energy-efficiency in data centers. We are in search of highly motivated students, with either a hardware or software background, that wish to collaborate with a research group with international projection.

Interested students may contact any of the group members, indicating which project they are interested in and attaching their CV.

The projects currently offered are the following:


Trabajos Fin de Grado (TFG)

Development of Dashboards and Visualization tools to show Key Performance Indicators in Data Centers

The goal of this project is to develop visualization tools that allow to show the KPI (Key Performance Indicators) of Data Centers as well as the variables that help exhibiting the thermal and energy inefficiencies. We are currently offering 3 different projects:

  • The development of 3D interactive thermal maps to show the hot-spots and thermal inefficiencies of the data center

  • The development of reconfigurable dasboard to show the Data Center KPIs, by using modern libraries such as 3Djs

  • The visual improvement of the tools to gather and store data about the Data Center inventory.

Contact Marina Zapater


Trabajos Fin de Máster (TFM) / Proyectos Fin de Carrera (PFC) / MSc Thesis


Automatic localization and inventory of Data Center assets

The goal of this project is to automatically identify and locate the different assets that can be found in a Data Center. Asset location comprises two different tasks:

  • The development of an NFC-based solution to tag and automatically identify the servers installed in a rack and update the data center inventory.

  • The localization of the nodes of a Wireless Sensor Network deployed at the data center facility to monitor the data room environment.

Contact Marina Zapater

Development of tools for Energy-aware Resource Management in Data Centers

Resource management is a well known concept in the Data Center world and refers to the allocation in a spatio-temporal way of the workload to be executed in the data center, optimizing a particular goal. Traditionally, these techniques have focused on maximizing performance by assigning tasks to computational resources in the most efficient way. However, the increasing energy demand of Data Center facilities has shifted the optimization goals towards maximizing energy efficiency.  Our research group has developed several algorithms that target the energy reduction in the resource manager.

The goal of this project is to develop software tools that enable to test this algorithms in a production Data Center. To this end, this project proposes the development of a plugin for the open-source resource manager SLURM.

Contact Marina Zapater

Anomaly Detection in Data Centers via Reputation Systems

Reliability is one of the key performance factors in Data Centers. The out-of-scale energy costs of these facilities, lead Data Centers operators to increase the ambient temperature of the data room to decrease cooling costs. However, increasing ambient temperature results in higher possibility of anomalous events. Thus, thermal anomalies in the Data Center need to be detected as soon as possible to optimize cooling efficiency and mitigate the harmful effects over servers.

This is a research-oriented project so, depending on the results of the project, the publication of the outcomes in relevant conferences or journals in the area is envisioned.

Contact Marina Zapater

GreenComputing techniques to improve the energy efficiency of software applications

The goal of this project is to explore the various software optimizations that can be applied to software in order to improve the energy consumption of an application without performance degradation, that is, without decreasing execution time. To accomplish this, different optimizations will be applied to high-performance benchmarks such as the SPEC CPU 2006 benchmark suite.

Contact Pedro Malagón

Runtime reconfiguration of Virtual Machines for improving energy efficiency in data centers.

One of the most common techniques applied in current cloud computing data centers is virtualization. KVM is one of the most common virtualization solutions (similar to others such as VMWare, Xen, etc.) and is based on QEMU virtual machine. The goal of this project is to substitute the QEMU binary translator by LLVM in order to be able to reconfigure the virtual machines (that is, changing the assigned resources) of a data center on runtime and thus obtain energy savings.

Contact José M. Moya