Data Center Management

• A data center is a physical facility housing servers, storage, network, and communication equipment for IT operations.
• It centralizes IT operations for storing, processing, and distributing data across an organization.
• Data centers provide access to large amounts of data and applications.

Data Center Components

• Core components: Hardware and software for storing data, including servers, routers, racks, and network infrastructure.
• Facility: Optimizes space and provides a temperature-controlled environment.
• Support infrastructure: Includes redundant power, physical security, and environmental controls.
• Operations staff: Oversee and monitor data center operations, ensuring maintenance and troubleshooting.

Types of Data Centers

• Enterprise: Deployed by an organization for its own use.
• Colocation: Rented physical space or racks within a facility.
• Cloud-based: Operate on a pay-as-you-go pricing model.
• Managed services: Outsourced data centers where management is handled by a third party.

Enterprise Data Center: Advantages

• Dedicated infrastructure for application deployment.
• Better control over data due to organizational ownership.
• Easy accessibility for troubleshooting.

Enterprise Data Center: Disadvantages

• Requires extra capacity for peak load requirements.
• Requires additional resources for continuous power and cooling.
• Consumes a lot of space and increases costs.

Cloud-Based Data Centers

• Customer is responsible for application security, OS, firewalls, accounts, and network.

Managed Services Data Centers

• Organizations lease physical access and outsource data center operations to a third party.
• Dedicated infrastructure exists for application deployment.

Managed Services Data Center: Advantages

• Organizations do not need to allocate resources for maintenance.
• Infrastructure maintenance is handled by the data center.
• Managed Services Data Center: Disadvantages
• Extra capacity and additional resources for power and cooling are required.

Data Center Role

• Houses critical and proprietary assets (data).
• Supports business applications.
• Provides data management services like backup and recovery.

Data Center Goal

• Provide 24/7 reliable services.
• Ensure flexibility and scalability.
• Offer long-term sustainability.

Enterprise Data Center Implementation

• Organizations select the site, customize the design and manage day-to-day operations.

Colocation Data Center

• Organizations rent physical space or racks.
• Customers manage their servers and storage.
• The colocation provider maintains building infrastructure and utilities.

Colocation Data Center: Advantages

• Cost-effective as multiple organizations share space and utilities.
• Scalable infrastructure enables quick scaling.
• Improved system reliability due to redundancies in utilities, security, and ISPs.

Colocation Data Center: Disadvantages

• May have impeded access due to facility rules and regulations.

Cloud-Based Data Center: Advantages

• Servers deployed and scaled within minutes.
• Better end-client experience due to global server availability.
• Customers avoid IT infrastructure maintenance responsibilities.

Cloud-Based Data Center: Disadvantages

• Relies on internet connectivity, which can be lost.
• Regulatory requirements may restrict data storage regions.
• Recurring monthly costs can exceed upfront costs.

Redundant Systems and Infrastructure

• Redundant cooling and power systems are critical for data center availability.
• Systems describe redundancy as N+1, 2N, and 2N+1.
• N+1 Redundancy: Components (N) have one independent backup (+1).
• 2N Redundancy: Components (N) each have an independent backup (2).
• 2N+1 Redundancy: Components (N) have an independent backup (2) and a single backup for each pair.

Racks

• Critical component of physical infrastructure.
• Constructed using steel frames and electronic structures.
• They improve performance and combat system failures.

Fault Tolerance

• Data centers must operate continuously.
• Fault-tolerant systems provide total availability by working uninterrupted even with component failures.
• A UPS provides 5-20 minutes of power during outages, sized for "peak" load, and continuously conditions power.
• Generators provide the power during extended outages and require maintenance considerations, fuel contracts, and scheduled tests during the design phase.

Fire Detection and Suppression

• Examples of Fire Detection:Smoke and heat alarms
• Fire Suppression:Automated systems, fire retardants, and sprinkler systems.

Location Considerations

• Lower real estate costs, adequate labor.
• Feasibility of reliable power and communication infrastructure.
• Proximity to stakeholders and vendors.
• Highway access, adequate parking.
• Water and fuel storage.
• Minimal risks of weather and natural disasters.
• Public safety.

Availability Measurement

• Mean Time Between Failures (MTBF) is the expected time between failures.
• Mean Time To Repair (MTTR) is the average time to repair a failed component or facility; redundancy duplicates systems or components.

Redundancy and Maintenance

• Redundancy avoids downtime during component failures.
• Proper maintenance is crucial for stable power availability.
• Types of Maintenance:Preventive, predictive, and scheduled.

Network Critical Physical Infrastructure

• NCPI is the layer below the physical infrastructure, crucial for running a data center successfully.
• The seven elements of NCPI: power, cooling, racks, cabling, security, management, and services.

NCPI: Power

• Reliable power infrastructure supports electrical services from the building entrance to the data center loads.
• NCPI power equipment includes generators, transformers, PDUs, power distribution panels.

Improving Data Centers

• Improve availability and reliability.
• Optimize Mean Time Between Failures (MTBF).
• Improve MTTR.
• Contain total cost of operations.

Data Center Availability Definition

• Readiness of systems to be accessible when needed.
• The probability that a system runs and is operational at a specific point.
• Can be increased by incorporating standby redundant systems.

IT Connectivity

• Connectivity: How well devices communicate with each other.
• Networks: Provide a common map for servers and equipment.
• Provides a means for data exchange

Improving Power Availability

• Utilize backup systems, such as generators with fuel.
• Have backup fuel in addition to what is stored.

Improving Cooling Availability

• Provide 2N redundant cooling capacity.
• Link to backup power systems that are quick to start after an outage.
• Rack availability requires considering future capacity increase.
• Ensure proper use of rack space.

Cable Availability

• Follow guidelines suggested by international standards using ANSI and TIA
• The ANSI/TIA-942 Telecommunications Infrastructure Standard provides design specifications for data center infrastructure.
• IEEE provides technical expertise in electronics and electrical engineering.

Key Environments

• Environments for operating a data center.
• Power Arrangement: Sufficient and redundant.
• Cooling Setup: Adequate.
• IT Connectivity: Proper.
• Space: Enough for the number of servers

IT Connectivity: Network Components

• Includes severs, switches, routers, controllers, and gateways
• Important to have redundant fiber connections with multiple internet service providers (ISPs).
• To help ensure the data center is always connected to the internet.

IT Connectivity Protocols

• Technologies for data center networking.
• Transport Control Protocol/Internet Protocol (TCP/IP)
• Ethernet
• VXLAN supports overlay network for virtualization.
• OpenFlow underlies software-defined networking (SDN). It helps determine data flow paths.

IT Connectivity: Development

• SDN efficiently accommodates varying workloads through data and control planes.
• Cooling removes pollutants by localized cooling.
• Fire protection systems have alarm capability, monitor, authorize, and restrict access to data centers.

Air Conditioning Specs

• Maintained: Ambient temperature, relative humidity and static pressure.
• Ensured: Airflow throughout racks with no leakage, spot cooling, and multiple HVAC systems (no centralized).

Raised Floor

• 12, 18, and 24-inch floors facilitate airflow and cooling depending on area size. This is more competitive than vinyl tile installations in terms of cost.
• Raised floors provide adequate cool airflow for easy maintenance.
• The height depends upon the size of the facility.

Data Center Planning

• Undertaken by qualified resources.
• Focused on spare spaces for core infrastructure.
• Plan for expansion in height and spare areas.

Data Center Architecture

• Must have: A single story with large floor plates, high ceilings, and few windows.
• Implement: Redundant power and communication services in highly secure areas.

Data Center Availability

• Is affected by power failures, equipment failures, network outages and can be measured by uptime and downtime.
• Formula: Availability = Uptime / (Uptime + Downtime)

Component Availability

• Formula: Availability = MTBF / (MTBF + MTTR)
  • MTBF = Mean time between failures = (Total uptime) / (number of breakdowns).
  • MTTR = Mean Time to Repair = (Total downtime) / (number of breakdowns).

Tier Classification

• The Uptime Institute's Data Centre Site Infrastructure Tier Standard provides an overview of data center availability.
• It also addresses the requirements for supporting critical systems.

Tier 1: Basic Site Infrastructure (N)

• Has a single distribution path for power and cooling with no redundant capacity components
• Is suseptible to disruptions when there are both planned and unplanned activities.
• Availability is 99.671%.
• Requires UPS, Cooling System and Generator

Tier 2: Redundant Capacity Components (N+1)

• Single path for power and cooling distribution
• Has: Redundant components and is slightly less susceptible to disruptions
• Provides:99.741 percent availability, and maintenace requires a shutdown

Maintaining Racks, and Fire Protection

• Can cause acute failure the equipment or heavy financial loss if racks aren't correctly maintained.
• Fire and Security systems are required to protect the data center from unauthorized access or a fire
• Cabling Infrastructure supports all networking and power cables, common problems include wrong plug-in cabling due to human error and overheating
• Building Management Systems (BMS) are NCPI elements.

Achieving Maximum Benefit

• The NCPI elements should be:
  • Designed for: Maximum availability and reliability
  • Regularly maintained for: Increased MTBF
  • Reduced Operational Dpowntime
  • Lows Costs (including Energy and Operating Costs)

Data Center Spaces

• Has gray spaces and white spaces which need efficiency to achieve better throughput and performance.
• Gray spaces are backend support systems that monitor the white space from within their confines
• Key environments for optimal conditions include cooling and power.

Power Supply Facilitation

• Is provided within a data center through the power grid (utility company), step-down transformers, back-up generators, and UPS systems.
• Transformers ensure correct voltage and current.
• MDB transfers power from grid/generator to equipment.

Cooling Technologies and Applictions

• Chilled water systems and CVC are cooling solutions.
• Cold aisle is a technology that facilitates cooling that is efficient for different data centers
• Designing a data center requires proper equipment necessary for undisturbed electrical supply coupled with adequate cooling systems.

Availability of Management and Services

• Ensured by maintained a backup procedure and process together with failover to maintain availabilty of the service,
• Maintaining a workforce thats properly trained and consistent with the latest industry requirements.

Organizing a Space

• A reliable, well-coordinated communication must occur between the IT equipment and backend support in the data center.
• Space may be physically expanded to accommodate the IT hardware, but ideally the demand must be offset in the existing equipment to be operationally efficient.

Increasing the White Space and Maximizing Profit.

• Utilizing Virtualization software on existing hardware
• By utilizing only efficient space designs.

Cables for Power

• The cabling network should be designed so that current needs and future power demands are met Proper temperature control and natural disaster-proofing should be incorporated.
• Cabinets and shelves should be compliant with the overall specifications that are needed by the data center.

Tier III: Concurrently Maintainable (N+1)

• Multiple power and cooling paths
• Allows Planned Maintaince Availability - 99.982%

Tier IV: Fault Tolerant 2N

• Multiple distribution and redundant paths that are Active.
• 99.995% availability.
• Infrastructure is sustailable even if unplanned or failed.

Data Center Resilience

• Achieved through parameters which includes Recovery Time Objective (RTO) to minimize business loss after failure
• Maintain backup equipment to ensure smaller RTO
• Includes Recovery Point Object (RPO): targeted amount of time where data loss is acceptable.

Data Center Spaces - Analysing and Increasing Use

• Increase the use of white space by virtualization software
• Maximized with proper organization during the design phase
• Cable and rack requirements must be factored for overall need according to heat and design

Data Center Operations - Challenges and Solutions

• Keep operational loads at or below limit
• Asses impact
• Asses failure impacts.

Data Center Security - Types

• Protection of the facility from theft or physical damage.
• Data security
• network security
• physical security

Preventive Security Options - Data Security

• Requires Standards and protocols set for proper data protection.

Preventive Security Options - Network Security

• requires Proper firewall and networking security protections.

Preventive Security Options - Physical Security

• Personnel must strike a perfect blend of technology, automation, and human intervention.
• Security policies set an implementation
• Security Layers protect at each access controlled gate
• Security protocols that feature layered security

Access Control

• Manage access lists to allow authorized personnel only, manage points and limit access and conduct biometric access measures.
• RFID Alerts asset tampering or need for Maintenace.

Questions cover data center accessibility, cable maintenance, availability metrics, and redundancy. Focus on network-critical physical infrastructure (NCPI) and proactive measures like redundant fiber connections and diesel generators.