If you are a System or Server Administrator, High-Availability or Business Continuity is the word you constantly hear from the management.
One of the most standard questions coming to our mind when we talk about High-Availability is what this High-Availability is? What is the percentage need to achieve that?
The word itself conveys you the answer that system or application needs to be available for all users and stakeholders at all time. But in the practical scenario, it is hard to maintain 100% uptime for many applications or servers at some points. That’s the reason SLA comes to the picture for hosting or co-location service providers. This means all the systems need to be thoroughly tested and carefully maintain with redundant components to ensure continued operations.
Let’s look for the major challenges System/Server administrator face in Server Room or Datacenter which can interrupt the operations or service.
- Power Outage
- Equipment Failure
- Human Interaction/Error
- Natural Disaster (Fire, Flood etc.)
To achieve High-Availability in Server Room or Datacenter follows the lifecycle of IT Infrastructure Management.
Four Major system components which can help achieve high-Availability or Business Continuity.
- Reliable or Redundant systems and components
- Backup and Recovery Implementation with Disaster Recovery Solutions.
- Define process management like Automation, Access right management, and change management.
- Advance monitoring and detection system.
As we speak about IT infrastructure many factors come in the picture. Most of them are physical components like Power, Network Connectivity, Servers, cooling systems, Fire Protections, moisture detections. Redundant system implementation helps you prevent a single point of failure.
Power
Network Connectivity
Servers
Carefully maintained servers and following service lifecycle for upgrade and replacement. This process helps to maintain system configuration at its best and help prevent physical failure of servers.
Cooling System
it is as critical as cooling systems as we all know Computer system, Servers generate lots of heat in case of cooling system failure at some point you require this system to avoid catastrophic data loss event. If you are working in Manufacturing or Production Company you might already have Fire and safety professional working. You can ask help from experts when implementing Fire detection and protection system. Or you can hire outsource the firm which provides such solutions.
Implementing such a system for 100 % redundancies comes with a cost. So effective planning help reduce cost, as well as help, maintain High-Availability.
Backup and recovery planning and implementation remain the key point to support the high-availability event if any failure occurs. For example, in case of a power outage, if you have, power generation system in place it helps to maintain your usage. Same way advance and smart UPS also help maintain load and power usage so in case of emergency or power failure it helps to prevent major data loss.
Backup solutions for systems and software carry out a leading role here. This helps in the event of data or system failure or accidental human interactions with production systems. If you maintain Backup policy and restore process in place, it helps to restore data at its previous stage. This helps the recovery of data at major data loss event for many critical business applications.
You can use any of enterprise backup solutions to backup system configuration and production data like Symantec net backup, Veritas or IBM Tivoli. If you are working with small or medium industry and cost or budget is an issue. You can always use Veeam Free backup solutions. Which help you to take windows system and server backup to USB or Shared network drive.
Disaster and Recovery Solutions: Disaster and Recovery planning required input from corporate management as well because of the cost and operations factor. Implementation cost will vary with the type of disaster and recovery with availability scenario. You can always use an in-house backup off-loading system to near site storage and by putting replication server to near or far operational site as a recovery plan.
One of the critical scenarios in Server and Datacenter Management is Process. This helps define 5 W’s of Datacenter Management what who why when where. So to define this process, you need implementing Access Management and Change Management.
Access management covers Access and Rights for the stakeholder or developer to physical as well as system access rights. Change Management help to identify the root cause for error in time of failure events.
Automation will help a completing task faster and safer as it minimizes human interaction in defining the process. As in IT, there are tons of examples where due to human error production data or file server data got deleted. This might help to avoid such incidents.
Advance Monitoring system help to prevent major system failure by giving alert in many ways. A monitoring system for power, network, system, temperature, humidity, smoke/fire detection will alert in before a time that helps to control failure event. Smart and advanced UPS and PDU come with smart monitoring tools. If any power failure or overload occurs, the system will notify the support person by email or text. Datacenter management tools can be integrated with temperature and humidity control sensors which detect any issue and report it to the system manager.
For system and network monitoring, many applications and tools are available in the market. System/Server Administrator can define the monitoring threshold for notification like Storage drive or Memory utilization. Many enterprise software and Open source available in the market to support monitoring systems. Few examples Microsoft SCCM, Wireshark, PRTG are paid software system admin can use. Nagios, Cacti is one of the best open source tools available in market system admin can use.
To maintain desired uptime placing the best equipment available in the market can’t guarantee High-Availability. To achieve High-Availability in the server room or data center requires redundant designs, right configuration of backup and disaster recovery solutions, process management and advanced monitoring systems.
In this article, we included the points System /Server Administrator has to consider while implementing or maintaining Server Room or Datacenter.
Hope this will help to discover critical factors of data center infrastructure.
Thank you for reading.
Mahesh ChandPosted Nov 22, 2018, 11:09 AM
This is awesome, Krunal. We don't have much content on this kind of details on infra or data centers. Thank you!