Professional Maintenance of the Data Center Infrastructure
Having a highly complex infrastructure, e.g. data centers and server rooms, IT security can only be guaranteed by regular maintenance. Even the failure of a single system component can lead to a chain reaction, which could bring an entire data center to a standstill in the worst case. A continuous surveillance of the different components and their performance can easily show failures. Hence, error sources can be eliminated more quickly and downtimes kept to a minimum. Since every data center is very individual, a customized service is essential. We offer IT services that data center operators can hardly manage themselves, at least not without a lot of effort. And we provide you with just the maintenance service that is necessary and reasonable.
What exactly does inspection and maintenance mean?
Inspection
Inspection involves a thorough review of the system or establishment. The current state is recorded and documented and performance, settings and data are checked. The aim of the inspection is to examine a system or establishment regarding a proper, secure and, hand in hand with that, flawless state. We check the different parts or units for signs of wear, and analyze the state of the working fund. On top of that, there is a visual examination of all relevant wearing parts, incl. measuring of attrition and corrosion tests. Based on that, the current and target state are compared and, if necessary, reasons for the signs of wear are determined and further consequences defined. The successful inspection is being tagged with an inspection plate and indicates the next date of inspection. Being part of maintenance, the inspection is conducted by our professional staff and an inspector. The inspection takes place in regular, previously defined intervals.
maintenance
Maintenance includes all activities that are intended to ensure the functionality of a system or machine. This includes inspection, maintenance and repair, as well as operational optimization and checking for weak points. The data center services are based on DIN 31051, which regulates the maintenance process by standard: “Maintenance is the combination of all technical and administrative measures as well as management measures during the life cycle of an observation unit to determine and assess the current situation -State as well as to maintain the functional state or the return to this, so that it can fulfill the required function.
Operation and maintenance of your data center
RZservices provides you with a service package that is specially tailored to your requirements and the nature of your data center. To serve z. B. our individual service level agreements. In addition, you can order a complete package as well as make use of partial services. Regardless of whether your data center is being planned or already in operation - DC-as-a-Service can be implemented at any time. Contact us!
Operations management
The longest phase in a data center's life cycle is undoubtedly the operational phase. Here it is important to maintain the normal operation of the data center and to operate the structural and technical systems in an optimal condition. For this purpose, services must be provided that are coordinated and sustainable. Maintenance measures in accordance with DIN 31051, but also individual maintenance strategies that offer the highest level of safety and efficiency, are essential for this. An overview of the infrastructure components used and their configuration are also included here, as well as the capacity of the data center.
Inventory and configuration management
Inventory and configuration management
Different trades and different technologies are used in the data center. In addition, the possibilities to program and parameterize systems are continuously increasing. Therefore, all technical systems and their configuration should be listed in a CMDB (Configuration Management Database), regularly adjusted and monitored. In addition to the inventory and the configuration, this also includes the documentation of the inventory.
The request:
Adoption of all configuration units and introduction of a CMDB. Structure, maintenance and responsibility for the system must be guaranteed.
Das Ergebnis:
Establishing and maintaining a CMDB in relation to the physical infrastructure is an important service provided by RZservices. For this purpose, we prefer to use the DC monitoring of our sister company RZproductsein, which also provides the inventory and documentation function. Operating parameters are visualized online and the inventory and configuration parameters are updated for every maintenance measure.
Maintenance strategies
Maintenance strategies
Basically, a distinction is made in the maintenance of the basic measures maintenance, inspection, repair and improvement. All measures serve the purpose of avoiding the wear limit, i.e. the failure of one or more systems. For this purpose, strategies and measures tailored to the systems and their use must be implemented. However, one should not underestimate the special features of a data center. What is meant are the two hearts that beat here in the form of IT and the physical infrastructure. All measures must not lead to any negative interaction or unplanned failures, because, for example, the repair of a system means at least the loss of redundancy. This requires a high level of experience and expertise in both disciplines.
The request:
Use of classic and individual maintenance strategies to improve operational safety and reduce costs in repairs. Compliance with standardized and manufacturer-specific requirements. Avoidance of interactions and unplanned failures.
The result:
All maintenance measures are carefully planned and prepared. Depending on the requirements for availability, redundancy concepts and equipment of the systems, a distinction is made between preventive and predictive maintenance. Depending on the scope of the measures, an experienced coordinator is deployed who has the overall function, i.e. the two hearts, in mind. The use of maintenance software (DC monitoring) is also ideal here. In addition to appointments, technical documentation, etc., important information about the history of the systems can be stored here, which simplifies diagnoses and repair measures. Such solutions also provide important information about the condition of the system between maintenance appointments.
Capacity management
Capacity management
The capacity of a data center in terms of size and number of racks is quickly identified. The situation is different with the capacity of the energy or cooling supply. Furthermore, the redundancies in the supply technology must also be taken into account. There is a close correlation between the degree of redundancy, the capacity provided and the capacity used. If, for example, two redundant UPS systems are operated above 50% of their capacity, the required redundancy may not be required in the event of a system failure. This must be continuously monitored and remedial action taken if necessary. But loads that are too low or an asymmetry in the load distribution can also have negative consequences for the security and efficiency of the data center.
The request:
Creation of transparency in the total capacity, the provided and used capacity. Monitoring of redundancy threats and asymmetrical load distribution.
The result:
Here, RZservices does a demand-oriented inspections of the load distribution in the supply technology. For this purpose, existing measuring devices are read, and your own uninterrupted measurements are carried out and documented or recommendations for action are given. This relatively complex method of inspection should, where possible, be replaced by the use of DC monitoring. It is possible to automate this monitoring as far as possible using a suitable measurement concept, so that the effort involved is minimized and the desired effect is still retained.
The highest demands are placed on the availability, security and economy of the data centers. With DC-as-a-Service, products and services are available to meet all requirements.
The steadily increasing legal and regulatory requirements for IT operations represent an enormous challenge for all IT areas. For this reason, more and more data center operators are deciding to focus on core competencies and tasks relating to maintenance, Passing on security and economic optimization to service partners. Due to the individuality of the various data centers and customer requirements, RZservices has developed the service portfolio DC-as-a-Service (Datacenter-as-a-Service).
DC-as-a-Service includes a number of classic, individual and adaptable models that cover maintenance and service from sub-areas to the entire data center infrastructure. All DC-as-a-Service activities are based on three pillars: availability, security and cost-effectiveness. We offer flexible service packages to support and optimize infrastructures, but also basic services such as maintenance, inspections, cleaning and relocations of data centers and server rooms.
RZservices provides you with a service package that is specially tailored to your requirements and the nature of your data center. To serve z. B. our individual service level agreements. In addition, you can order a complete package as well as partial services. Regardless of whether your data center is being planned or already in operation - DC-as-a-Service can be implemented at any time. Contact us!
Our services
Service Hotline
Service Hotline
In order to be able to react appropriately and professionally to operational disruptions, it is essential that a knowledgeable service provider can be reached by telephone. Only in this way can suitable measures be taken to restore the original operation of the systems and components.
The request:
The service provider is on call 24 hours a day, 365/366 days via a service hotline
The result:
The conclusion of a service contract already includes the basic flat rate of around-the-clock on-call service. After receipt of the fault report, a technician will respond on the following working day during normal working hours
NOC (Network Operations Center)
NOC (Network Operations Center)
In addition to the telephone service hotline, the technical detection of operating and malfunction messages, as well as the forwarding of this information to a permanently manned office, is a higher level of operational management. In this way, technical information on all systems and components can be transmitted and processed without the involvement of human resources.
The request:
Personal and technical transmission, as well as defined elimination of operating and malfunction messages.
The result:
Telephone availability around the clock and use of a monitoring system. Mechanisms agreed in advance are initiated here, recorded and processed using the ticket system. The RZservices Network Operations Center (NOC) processes every malfunction until a ticket can be closed again.
Performance and redundancy test
Performance and redundancy test
Does the performance and reliability of the data center keep what the planning promises? A comprehensive load and redundancy test should be a matter of course, especially with newly built data centers. Because the time before the IT components are settled and the associated start of productive operation of the data center is usually the only and last opportunity to test the promised properties in terms of performance and redundancy. This is the only way to simulate and document the later possible operating states using special test scenarios.
The request:
Creation of test scenarios as a simulation of the various possible operating states. Methods for checking the various redundancy concepts. Documentation of the conditions provoked by the test scenarios.
The result:
Die RZservices ist spezialisiert auf diese Art der Performance- und Redundanztests. Als Mindestvoraussetzung sind Prüfszenarien, welche die Versorgungstechnik in einem sehr wenig belasteten Zustand betreiben, in verschiedenen Teillastbereichen bis hin zur Volllast. Auch kommt es in diesen Szenarien gezielt zu Ausfällen von Versorgungsanlagen. Als eigentliche Last kommen Lastbänke in den Racks zum Einsatz. Diese Lastbänke erzeugen sterben Wärmelast ähnlich wie es später sterben IT-Komponenten tun. Zur Dokumentation der Prüfzenarien dienen Checklisten die mit Messdaten aus verschiedenen Messgeräten ausgefüllt werden. Besser noch ist der Einsatz eines DC-Monitoring. Hier can dann sterben verschiedenen Szenarien auch grafisch dargestellt und dokumentiert werden.
SLA (Service Level Agreement)
SLA (Service Level Agreement)
Depending on the requirements for the availability of the data center infrastructure, special conditions should also be created for sensitive areas. It is often not only enough to be available, but also to be on site as quickly as possible in order to be able to eliminate the cause of the malfunction
The request:
Location-based and individual services. Modular and trade-related response times.
The result:
Development of a customer-oriented service level agreement as part of the entire service contract. Clearly defined and assured response times for technicians.
Our references
-
EDEKA: more security for IT
The two builders EDEKA Südwest and EDEKA Nordbayern-Sachsen-Thüringen were looking for suitable locations for their IT and commissioned the DATA CENTER GROUP as general transfaree and project partner to plan and build two data centers.
-
KUKA AG: Green and Autarchic
KUKA and DATA CENTER GROUP develop a Data Center with regards to digitalization
-
WESTFLEISCH: 8 modular DC-IT containers
Construction of 8 modular IT outdoor containers as general contractor
© DC-Datacenter-Group GmbH
Phone +49 2741 9321-0