Understanding Planned and Unplanned Tasks
In the fast-paced and ever-evolving field of information technology, particularly in hosting services, the effective management of tasks is crucial for maintaining service quality and ensuring customer satisfaction. This article delves into the intricacies of managing planned and unplanned tasks within a hosting company, focusing on the roles and responsibilities of system administrators (sysadmins).

We will explore the distinction between planned tasks, such as routine maintenance activities, and unplanned tasks, which typically involve responding to incidents or alerts. By establishing a comprehensive framework for organizing and prioritizing these tasks, hosting companies can enhance their operational efficiency and preparedness for unexpected challenges. Through a practical example, we will demonstrate how a sysadmin team can effectively balance these responsibilities, ensuring that both routine maintenance and emergency incidents are handled with optimal resource allocation and prioritization.
Planned Tasks (Maintenance)
Planned tasks, or maintenance tasks, are those that can be scheduled in advance. This includes software updates, security checks, and other recurring tasks necessary for server maintenance.
| In a hosting company, a sysadmin schedules a software update requiring 1 to 3 hours. This maintenance is set to be carried out weekly. |
Unplanned Tasks (Incidents)
Unplanned tasks arise unexpectedly, such as incidents or alerts generated by services or reported by clients.
| A client reports a website downtime hosted on one of the company’s servers. This incident was not planned and requires immediate attention. |
Work Organization
Creating a Maintenance Catalog
We first create a catalog that includes all planned maintenance tasks, their estimated duration, and frequency.
Table 1: Maintenance Catalog
| Task | Estimated Duration | Frequency |
|---|---|---|
| Software Update | 1-3 hours | Weekly |
| Security Checks | 2 hours | Daily |
Daily Resource Table for Maintenance
The following table exemplifies how to organize the maintenance catalog to calculate daily human resource requirements.
| Task | Estimated Duration | Frequency | Resources per Day |
|---|---|---|---|
| Software Update | 1-3 hours | Weekly (1 time) | 0.5 person |
| Security Checks | 2 hours | Daily | 2 persons |
| Data Backup | 1 hour | Daily | 1 person |
| Server Optimization | 4 hours | Monthly (4 times) | 0.2 person |
| Log Review | 1.5 hours | Daily | 1.5 persons |
Note: The «Resources per Day» column is calculated by dividing the total estimated duration of each task (adjusted for its weekly or monthly frequency) by the number of working hours in a day.
Resource Allocation for Unplanned Tasks
Based on the history of incidents, we allocate resources to handle unplanned tasks, balancing them with planned tasks.
| If historically 2 critical incidents are reported per day, we assign at least 1 sysadmin exclusively to these tasks, with the ability to scale up to 2 or 3 sysadmins during peak activity. |
Priority Queues
We establish a priority queue for sysadmins assigned to maintenance, indicating who should assist in incidents if the demand arises.
Practical Example of Organization in a Sysadmin Team
Consider a team of 5 sysadmins in a hosting company:
- 1 sysadmin permanently assigned to incidents and alerts.
- 4 sysadmins performing planned maintenance tasks.
Workflow Diagram:
- Incidente Queue: It starts with the incident queue, where incidents are reported and handled by the specifically assigned Sysadmin 1.
- Maintenance Queue: Simultaneously, there exists a maintenance queue where Sysadmins 2, 3, 4, and 5 (identified in the diagram as «Sysadmins 2, 3, 4, 5 (Maintenance)») carry out maintenance tasks.
- Overload of the Incident Queue: If the incident queue exceeds a certain number of incidents (X), an escalation mechanism is activated. The first sysadmin from the maintenance queue leaves their tasks to assist in reducing the incident queue.
- Return to Maintenance: When the incident queue falls back below the threshold X, the sysadmin returns to their maintenance tasks.
- Additional Escalation: If the incident queue continues to increase and surpasses a higher threshold (X+Y), another sysadmin from the maintenance queue is brought in, and so forth recursively until the situation stabilizes.













