Power and cooling

Lately, the most discussed subject in datacenters is Power and Cooling. Why is this such a big issue all of a sudden?

Power bill

Usually a system administrator or a datacenter operator knows what a typical server cost. How many of them know what a kilowatt-hour (kWh) of energy costs? Do YOU know? And who pays that bill? The IT departyment?

The typical power consumption of a full-blown datacenter, including air-conditioning today is in the order of several megawatts. With a power price of 6 dollar cents per kWh (which is typically what large customers pay), the power bill of one megawatt is $1440 per day and $525,600 per megawatt per year!

In 1999, the price of a barrel oil was 11 dollar. Now, the price is 120 dollar. This means that the price of energy is rising at an incredible speed, and there is no end in sight. Therefore, the power bill will go up the coming years, and more attention should be paid to reducing the power consumption in datacenters.

It becomes now viable to place new datacenters in the neighborhood of “green” power plants, like water dams (the water can also be used to cool the datacenter). The cost of moving data through a fibre optics cable is much less than the cost of moving energy from a energy plant to a datacenter.

Lower consumption

For every dollar spent on servers, the same amount is spent on power. Not only the kWh´s of the server itself must be paid, but also on cooling the system, the cost of expanding the air-conditioning system and the uninterruptable power supply (UPS).

Therefore it is a good financial investment to try to keep the power consumption of servers low. A typical Intel CPU in 2005 used ten times more power than the same CPU capacity today. This brought the energy bill down considerably. Therefore, the focus for Intel these days is not to increase the speed of a CPU, but to lower the power consumption. The money saved in power, UPS systems and cooling could be invested in more (low power) CPU’s.

Virtualization is another method to decrease power consumption. Not only can resources be better utilized, and CPU’s used as efficiently as possible, but also the possibility to move running virtual machines to other physical machines (using for instance vMotion from VMware) can reduce hot spots in the datacenter.


This entry was posted on Friday 12 September 2008

Earlier articles

Infrastructure as code

My Book

DevOps for infrastructure

Infrastructure as a Service (IaaS)

(Hyper) Converged Infrastructure

Object storage

Software Defined Networking (SDN) and Network Function Virtualization (NFV)

Software Defined Storage (SDS)

What's the point of using Docker containers?

Identity and Access Management

Using user profiles to determine infrastructure load

Public wireless networks

Supercomputer architecture

Desktop virtualization

Stakeholder management

x86 platform architecture

Midrange systems architecture

Mainframe Architecture

Software Defined Data Center - SDDC

The Virtualization Model

What are concurrent users?

Performance and availability monitoring in levels

UX/UI has no business rules

Technical debt: a time related issue

Solution shaping workshops

Architecture life cycle

Project managers and architects

Using ArchiMate for describing infrastructures

Kruchten’s 4+1 views for solution architecture

The SEI stack of solution architecture frameworks

TOGAF and infrastructure architecture

The Zachman framework

An introduction to architecture frameworks

How to handle a Distributed Denial of Service (DDoS) attack

Architecture Principles

Views and viewpoints explained

Stakeholders and their concerns

Skills of a solution architect architect

Solution architects versus enterprise architects

Definition of IT Architecture

What is Big Data?

How to make your IT "Greener"

What is Cloud computing and IaaS?

Purchasing of IT infrastructure technologies and services

IDS/IPS systems

IP Protocol (IPv4) classes and subnets

Infrastructure Architecture - Course materials

Introduction to Bring Your Own Device (BYOD)

IT Infrastructure Architecture model

Fire prevention in the datacenter

Where to build your datacenter

Availability - Fall-back, hot site, warm site

Reliabilty of infrastructure components

Human factors in availability of systems

Business Continuity Management (BCM) and Disaster Recovery Plan (DRP)

Performance - Design for use

Performance concepts - Load balancing

Performance concepts - Scaling

Performance concept - Caching

Perceived performance

Ethical hacking

The first computers

Open group ITAC /Open CA Certification

Sjaak Laan


Recommended links

Ruth Malan
Gaudi site
Esther Barthel's site on virtualization
Eltjo Poort's site on architecture


Feeds

 
XML: RSS Feed 
XML: Atom Feed 


Disclaimer

The postings on this site are my opinions and do not necessarily represent CGI’s strategies, views or opinions.

 

Copyright Sjaak Laan