Monday, 22 October 2012

Impact of Downtime.

 Downtime must be considered from a number of aspects.
-         System outage time.
-         System/business recovery time.
o   Reputational integrity – directly affects share price.
o   Cost of customer loyalty
o   Potential gain for competition.
-         For financial services companies, other key considerations could include regulatory requirements
-         By understanding the total time lost to unscheduled downtime (outage + Recovery) x the total cost/hr to the business by not having system availability, the business shout be able to determine the level of redundancy required to support the business needs.
-         The potential losses incurred as a result of downtime must be balanced against the additional costs associated with higher levels of availability.
-         Fines imposed by regulatory authorities for late submissions of reports.
-         Penalties imposed for a lack of up to date trading information.
-         Can result in substantial financial loss due to inability to trade.

Business Needs Assessment.


-         How dependent is the business on the data center
-         What physical size/space is required
o   Now
o   Later
o   Ultimately
-         What level of availability is expected.
o   Variable?
-         What is the impact of downtime on the business in $
-         What can the business afford in $
-         What already exist, what other facilities could be utilized.
Three main are need to be considered:
1.       Available time for planned maintenance
-          Determined by type of business, workload and redundancy level
2.      Acceptable amount of time for unplanned outage
-          Determined by level of dependence of the business on the availability of the data center
3.      Overall impact of unscheduled outage on business performance
-          The effect of 2. Above on the business as a whole in terms of lost revenue, damage to reputation, exploitation of opportunity by competitors.

Important notes to be considered.
-         The choice of the most appropriate Data Center availability level for the business is a C class decision.
-         TCO and ROI must be fully understood.
-         The facility design must meet the business availability requirement.
-         All options must be explored.

Friday, 19 October 2012

Design and Project Process






Before a data center design decision can be taken, all the key stakeholders must be considered, these include:
-         The C Class people
-         Stakeholders
-         Customers
-         Staff
-         Suppliers
Client Brief:
-         IT capacity ramp up over asset life
-         Incremental while space fit out / modular concepts.
-         M&E Resiliency Requirements.
N+1 or 2N same of each compartment /room?
-         Telco Resiliency Requirements.
-         Technologies to be implemented.
-         PUE Target.

Planning Phase:
-         Consultants are appointed.
-         Crucial decisions are made.
-         The design intent document is established
-         The cost plan is determined
-         The project timeline is established
-         The client brief is signed off
-         Funding approval and budgets are agreed
-         Regulatory planning and codes consent are applied for
-         IT equipment delivery, Storage and installation plan is created.
Design Phase:
-         The main contractor is appointed
-         The entire project team is identified
-         The system to be commissioned are determined
-         An architectural review of room building is conducted
-         A need assessment /inventory of IT requirements is performed.
-         The design intent documentation is submitted
-         IT, facilities and architecture are reviewed.
-         The scope of work for all participants, including contractors and vendors is reviewed.
-         The utility providers are established
-         The commissioning plan is produced
-         The IT infrastructure and network design is agreed.

Build Phase
-         Appoint M&E Contractors.
-         Commissioning consultants define acceptance criteria
-         Milestone monitoring is performed
-         User acceptance testing at manufacture’s site
-         Pre-functional testing is completed as required.
-         Site inspections and progress reports are submitted.
-         Design modifications documented and approved
-         Change order process and approval authority are monitored
-         Suppliers train M&E Operational staff.
-         Commissioning phase.
Construction Phase Handover:
-         Equipment suppliers submit O&M documentation to main contractor.
-         M&E contractors’ staff carries out acceptance test in conjunction with contractors.
-         On site functional performance and integration testing.
-         Site /Equipment audit – M&E Asset list prepared.
-         Manufacturer warranty audit.
-         Health & Safety plan submitted including residual risks.
-         Final documentation submittal including ‘As built’ drawings and test results.
-         Change over temporary utility supplies to permanent.
Post Construction Handover – Client Handover Phase
-         O&M procedures established – Standard Operation Procedures SOP, Emergency Operation procedures EOP and Planned Maintenance Procedures PMP
-         IT equipment installed.
-         Document storage and document modification process defined.
-         M&E personnel testing and assessment
-         Moves, Adds and changes procedures established.
-         Change control procedures and policies implemented.

Sunday, 7 October 2012

Data Center Terminology Continued



Power and cooling density:
-         Power and cooling were defined as w/ft2 or W/m2 or British thermal units (BTU)/hr or KW/hr, but the true density should be determined by KW/rack.
-         High density is above 10KW/rack.
-         Medium density is from 4 KW/rack up to 10 KW/ rack.
-         Low density is up to 4 KW/rack
-         Previously the measurement of cooling was by Tons
-         Tons are related to the quantity of ice used to cool.
-         To change from Ton to Kwc. KWc= Ton x 3.516

Levels of redundancy:
-         N – is need and meets base requirements with no redundancy.
-         N+1 – is one additional unit/path/Module more than the base requirement.
-         2N – is two complete units/paths/modules of every one required for the system base.
-         2(N+1) – Two complete (N+1) units/module/paths if there is any failure of one system still leaves and entire stem still working.

Criticality Classification:           
-         N = Single path “No Redundancy”
-         N+1 = Single path with some redundancy
-         2N = Concurrently maintainable
-         2(N+1) = Concurrently maintainable with full fault tolerant-         

Availability rating:
99%                 annual time unavailable = 88 hours
99.9%             annual time unavailable = 8.8 hours
99.99%           annual time unavailable = 53 Minutes
99.999%         annual time unavailable = 5.3 minutes
99.9999%       annual time unavailable = 32 seconds

Uptime Institute Tier IV is 99.995% or 26 minutes/year

Metrics.

            The main power efficiency metrics currently in use within the dc industry are based on the “ratio of IT load to Facilities load”
-         This is referred to as the facility efficiency.
-         Two commonly used metrics are PUE & DCiE

Data center infrastructure efficiency
DCiE = (IT load) /  (Total facility load) x 100%

Power Usage Effectiveness
            PUE (Total facility load) (IT load)

What is the IT load?
            This relate to the power consumption of all of the IT equipment within the data center.

What is the facility load?
            This related to the mechanical and electrical systems that support the IT electrical and such as cooling systems (Chiller plant, fans, pumps) air conditioning units, UPS, PDUs etc.

PUE Version 2 has 4 Categories:
            PUE0 =This is still power (KW), but uses case measurements (highest peak measured)
            PUE1 =This is annualized energy measured at the output of the UPS (Lowest accuracy).
            PUE2 =This is annualized energy at the output of the PDU (better accuracy)
            PUE3 =This is annualized energy at the rack/input of IT equipment (best accuracy) 

Future metrics
CUE : Carbon usage effectiveness (Green Grid)
WUE : Water usage effectiveness (Green Grid)
CADE : Carbon average data center efficiency (Uptime inst.)
DPpE : Data Center performance per energy (Green IT promotion Council)

Saturday, 6 October 2012

Data Center Terminology


Important Definitions.


What is a Data Center?
TIA 942:
-         Data center is a building or portion of a building whose primary function is to house a computer room and its support areas.
EU CoC for data centers:
-         The term ‘Data Center’ includes all buildings, facilities and rooms which contain enterprise servers, server communications, cooling and power equipment and provide some from data service.
Efficiency:
-         The ratio of the output to the input of any system.
-         Skillfulness in avoiding wasted time and effort.
Effective:
-         Producing the intended or expected result.
Resilience:
-         It is the ability to provide and maintain an acceptable level of service in the face of faults and challenges to normal operations.
Redundancy:
-         The duplication of critical components of a system with the intention of increasing reliability of the system, usually in the case of a backup or fail-safe.
Availability:
-         The degree to which a system, subsystem, or equipment is operable and in a committable state of a mission, when the mission is called for an unknown, i.e., a random time.
Reliability:
-         The ability of a system or component to perform its required function under stated conditions for a specified period of time. It is often reported as a probability.

Friday, 5 October 2012

Data Center Standards and Regulations

What is the relevant industry bodies?

There a lot of companies which could be considered as relevant bodies following is list of them.

1- Uptime Institute which is founded in 1993
2- ASHRE  "American Society of heating, Refrigeration & air conditioning engineering, Founded at 1894
3- The green Grid Founded at 2007
4- AFCOM Association for computer operations management, founded 1980
5- TIA Telecommunication industry association, founded 19888
6- BICSI Building industry services international, founded 1974
7- DOE US department of Energy
9- EPA Environmental Protection Agency.

What is the standards and regulations for data center?

The current Standards and Reference Documents are listed as following.

1- Uptime Institute Tier ratings for data centers.
2- ASHRAE TC9.9 Guidelines for mission critical facilities
3- ANSI/TIA 942 - Infrastructure Standard for data centers
4- ANSI/BICSI 002 - Data Center design and implementation best practice.
5- EU-COC the EU code of conduct for data center.











Sunday, 30 September 2012

What is the history of data center?

The Beginning of the Computer Era - A Dedicated Room

             While the data center as we know it was perfected during the dot com boom of the late 1990s, data centers actually have their roots in the earliest beginnings of the computer era. Early computer systems, which were huge, room-sized machines, required a lot of space and a controlled environment. The complexity of operating and maintaining these machines also led to the practice of secluding them in dedicated rooms.

Computer security became a consideration during this era. These early computers were incredibly expensive, and many of them were used for military purposes or important civilian business ventures. A dedicated room allowed businesses and organizations to control access to the machine.
Another factor influencing the trend toward separate computer rooms was the need to keep systems cool. Early computer systems used a great deal of power and were prone to overheating. Dedicated rooms could be climate controlled to compensate for the tendency to overheat.
These early computers required a multitude of component-connecting cables, and these cables needed to be organized. This led to the creation of some of the data center standards we know today. Racks were devised to mount equipment, and cable trays were created. Also, floors were elevated to accommodate these early computers.

The Advent of Microcomputers

                  During the 1980s, the computer industry experienced the boom of the microcomputer era. In the excitement accompanying this boom, computers were installed everywhere, and little thought was given to the specific environmental and operating requirements of the machines.
Organization of information was difficult to achieve, and lost data became a major concern. Information technology teams were developed to maintain and install these early microcomputers, but clearly, the industry needed a solution.


The "Data Center" is created
Soon the complexity of information technology systems demanded a more controlled environment for IT systems. In the 1990s, client-server networking became an established standard. The servers for these systems began to find a home in the old dedicated computer rooms left from the early computers.
In addition to putting servers in a dedicated room, this time period saw the invention of the hierarchical design. This design came about through the easy accessibility of inexpensive networking equipment and industry standards for network cabling.
The term "data center" first gained popularity during this era. Data centers referenced rooms which were specially designed to house computers and were dedicated to that purpose.

The Internet Data Centre
As the dot com bubble grew, companies began to understand the importance of having an Internet presence. Establishing this presence required that companies have fast and reliable Internet connectivity. They also had to have the capability to operate 24 hours a day in order to deploy new systems
Soon, these new requirements resulted in the construction of extremely large data facilities. These facilities, called "Internet data centers" were responsible for the operation of computer systems within a company and the deployment of new systems. These large data centers revolutionized technologies and operating practices within the industry.
However, not all companies could afford to operate a huge Internet data center. The physical space, equipment requirements, and highly-trained staff made these large data centers extremely expensive and sometimes impractical.

Now - Private Data Centers, Improved Standards
Private data centers were born out of this need for an affordable Internet data center solution. Today's private data centers allow small businesses to have access to the benefits of the large Internet data centers without the expense of upkeep and the sacrifice of valuable physical space.
These days, operating and constructing data centers is a widely-recognized industry. New standards for documentation and system requirements add a high level of consistency to data center design. Disaster recovery plans and operational availability metrics ensure the reliability of today's data center systems.

What the Future Holds
The future of data center design is likely to reflect today's emphasis on green practices. Environmentally responsible computer and networking systems, as well as refined operation practice, are likely to shape the data centers of the future.
Managed Hosting provider delivers a "higher level" of managed IT services for deploying and hosting e-business, security, disaster recovery, and business continuity solutions for the mission-critical applications






What is the data center?
There are a lot of Data Center definitions
- A data center is a facility used to house computer systems and associated components, such as telecommunications and storage systems. It generally includes redundant or backup power supplies, redundant data communications connections, environmental controls (e.g., air conditioning, fire suppression) and security devices.

- A data center is a centralized repository, either physical or virtual, for the storage, management, and dissemination of data and information organized around a particular body of knowledge or pertaining to a particular business.