Published on February 26, 2014
A Cyber Physical Approach to Combined HW-SW Monitoring for Improving Energy Efficiency in Data Centers Josué Pagán, Marina Zapater, Oscar Cubo, Patricia Arroba, Vicente Martín and José M. Moya Universidad Politécnica de Madrid 1 / 20
Contents 1. Power consumption problem in Data Centers I. Introduction II. Related work 2. Optimization Framework & Data analysis I. A Cyber-physical system II. Data analysis and sensor configuration 3. Results 4. Conclusions 2
Data Center. Consumption 1. POWER CONSUMPTION PROBLEMS IN DATA CENTERS 3
1. Power consumption problems in Data Centers • The numbers of the energy problem: – DC world power consumption >1.3% – In urban areas >50% of DC exceeds power grid capacity – USA: 80 TWh/year in 2011 = 1.5 x NY Power >600 TWhr expected in 2015 in the global footprint • Data Centers’ power consumption is unsustainable Projection of total electricity use by datacenters in the US and the world based on Koomey’s and EPA’s data 4
1. Power consumption problems in Data Centers • Related work (approaches) Cooling • Allow the room temperature to increase • Longer task → cooler server • Balancing workload between servers Computation • Reducing voltage/ frequency (DVFS) – These two approaches are not enough individually Holistic (IT+cooling) • Room environment affects (environmental monitoring) • Measuring server, workload and environmental variables to improve energy efficiency → usage of a CPS 5
1. Consumption problems in Data Centers Requirements Energy optimization Our contributions Make a holistic optimization framework including environmental, server and workload information Dynamically adapt on runtime to workload and environment Gather, monitor and analyze in real time Gather useful data at the appropriate rate In a non-intrusive way, reducing the data collected with an adaptable sampling rate 6
Cyber-Physical System. Data acquisition 2. OPTIMIZATION FRAMEWORK & DATA ANALYSIS 7
2. Optimization Framework & Data analysis • One step ahead. Optimization – 80% Wpeak – 30% of workload (↓η) GATHER DATA PROPOSE OPTIMIZATIONS GENERATE KNOWLEDGE – An energy model supposes apply optimizations over the Data Center 8
2. Optimization Framework & Data analysis • Monitoring – How a Data Center works? – 30-50% cooling→ energy optimization 9
2. Optimization Framework & Data analysis • What measure and why – Environmental monitoring Inlet and outlet temperature Differential pressure – Server monitoring Server consumption, CPU temperature, fan speed • …to predict 10
2. Optimization Framework & Data analysis • How… • exploring sampling intervals – Temperature and power values for AMD server under the benchmark SPEC CPU 2006 – Different sampling rates for different parameters 11
2. Optimization Framework & Data analysis • Using… Multilevel star topology architecture WSN - Reconfigurable low -power: only useful data without information loss - Adapt to changes in the environment RM - Spatio-temporal allocation - Possibly to change decisions if needed Gateway -Fan-less, managed with a light OS -Receive, store, analyze and convert data. Establishes a timestamp. -Sends data to the opt. platform Server Sensors - Internal sensors - Polled via SW Air conditioning - Exhaust temperature, RH% and airflow 12
3. RESULTS 13
3. Results WSN deployment • Applied over Magerit Supercomputer in CeSViMa Supercomputing and Visualization Center of Madrid • Cluster 9 racks 260 servers 14
3. Results – The goal: develop techniques to allow energy optimization in real environments – With reconfigurable sampling rate: – we achieve up to 68 % of reduction in gathered data – Increase the WSN’ s life time depending on the occupancy 15
4. CONCLUSIONS 16
4. Conclusions Energy efficiency has to be faced in a holistic way We propose an optimization framework monitoring environmental, server and workload parameters After a first monitoring study: a WSN has been deployed to gather environmental data Up to a 68% of reduction in the amount of gathered data Maximizing the life time of WSN nodes Solution applied in a real case study 17
Thank you for your attention FIN This project has been funded in part by the INNPACTO LPCLOUD: "Optimal Management Of low-power modes in cloud computing" IPT2012-1041-430000, developed in collaboration with Elite Ermestel and Converging Technologies and the CDTI project CALEO: Distribution of operational thermal and optimization of energy consumption in data centers, "developed in collaboration with INCOTEC. The author gratefully acknowledges the computer resources, technical expertise and assistance provided by the Supercomputing and 18 Visualization Center of Madrid (CeSViMa).
4. Results and Conclusions • Results: gathering data • Inlet and outlet temperature 19
Magerit Supercomputer • Cluster 9 racks 260 servers 245 are IBM PS702 2S o 16 Power7 processors @ 3.3 GHz o 32 GB of RAM 15 are IBM HS22 o 8 Intel Xeon processors @ 2.5 GHz o 96 GB of RAM 200 TB of storage 20
Software • Pasarela 22
Cyber-Physical System 23
Psychrometric chart 24
Differential pressure and airflow 25
Installing nodes [6T+1H] 26
Presentación que realice en el Evento Nacional de Gobierno Abierto, realizado los ...
In this presentation we will describe our experience developing with a highly dyna...
Presentation to the LITA Forum 7th November 2014 Albuquerque, NM
Un recorrido por los cambios que nos generará el wearabletech en el futuro
Um paralelo entre as novidades & mercado em Wearable Computing e Tecnologias Assis...
What is ductal carcinoma in situ (DCIS)? Ductal carcinoma in situ (DCIS) is breast cancer, but it is non-invasive ... Posted February 28, 2013 ...
DCIS or Ductal Carcinoma In Situ is a non-invasive breast cancer discovered by breast screening mammography. In most cases DCIS does not progress to ...
All Topics → Forum: DCIS (Ductal Carcinoma In Situ) ... 2013 10:54AM Latest post: Feb 23, ... Recently diagnosed with DCIS and 6 months pregnant.
Version 2013.1D DCIS ist eine lokalisierte Erkrankung und sollte primär mit lokalen Maßnahmen behandelt werden.
Hallo Lindeli Ich habe meine DCIS-Diagnose letzten Juni 2013 erhalten; Rechte Brust DCIS high grade Grösse ca. 5,5cm (habe grosse Brüste Cup C/D)
DCIS - ahnungslos - plötzlich reingerutscht.. Strahlentherapie ... 20.09.2013, 18:56 Pfesel. Neuer Benutzer : Registriert seit: 19.09.2013.
DCIS 2016, November, 23rd-25th, Granada (Spain) Home; Call for Papers; Committees; History; Authors; Submission; Search. Search... You are here: Home. Home ...
Bereits im Februar 2013 hat Breast Cancer Action eine Überarbeitung des DCIS-Informationsblattes veröffentlicht. Da dieses nur in englischer Sprache ...
DCIS — Ductal Carcinoma In Situ Save as Favorite. Sign in to receive recommendations . Ductal carcinoma in situ (DCIS) is the most common type of non ...