Eay physical infrastructure management tools were limited in scope and required considerable human intervention. While they would warn that a particular parameter had been exceeded, the operator would have to determine what equipment was affected by the error. First-gen tools could not make correlations between a physical infrastructure device and a server, nor were they capable of initiating actions to prevent downtime, such as speeding up f to dissipate a hot spot.
Never management tools are designed to identify and resolve issues with minimum human intervention. By correlating power, cooling and space resources to individual servers (physical and virtual), DCIM tools today can proactively inform IT management systems of potential physical infrastructure problems and how they might impact specific IT loads. Newer planning software tools illustrate, through a graphical user interface, the current physical state of the data center and simulate the effect of future physical equipment adds, moves, and failures.
Data center architecture is the physical and logical layout of the resources and equipment within a data center facility.
It serves as a blueprint for designing and deploying a data center facility. It is a layered process which provides architectural guidelines in data center development.
Some features/functions to consider:
Open source — vendor-neutrality is key, as very few data centers are standardized on a single vendor from top to bottom. Open source DCIM software can integrate data from, for example, uninterruptible power supply (UPS) systems, power distribution units (PDUs), and cooling units from three (or more) different vendors.
In addition, with open protocols, it is quite easy to add additional software tools and expect them to communicate and work together effectively.
A data center is said to be carrier-neutral if a customer can order cross connections or communications services from any existing provider and the data center provider actively tries to court additional carriers into the facility.
Here are a few we see more often than we’d like:
No, DCIM tools consist of a collection of software applications (outlined above), data collection tools, and a dashboard. The data collection is generally done by devices like meters, power protection devices, embedded cards, programmable logic controllers (PLCs), and sensors, which gather data and forward it to management software for processing.
The other component of DCIM is a dashboard. Critical information from the DCIM software and data collection tools needs to be aggregated and presented so IT managers can visualize the data in a way that is meaningful and actionable. Dashboards can be configured for different needs, for instance to focus on the performance of the IT equipment versus the physical infrastructure (cooling, power, security).
A data center (or datacenter) is a facility composed of networked computers and storage that businesses or other organizations use to organize, process, store and disseminate large amounts of data. A business typically relies heavily upon the applications, services and data contained within a data center, making it a focal point and critical asset for everyday operations.
Monitoring and automation software can do things like:
Planning and implementation software can do things like:
Data center management refers to a small number of employees who have been designated and hired to manage large data sets and hardware systems that are usually part of a large distributed network. The data center is responsible for the management of significant amounts of data and the hardware required to store it and distribute it to users.
Data center management plays a crucial role in protecting data and keeping it secure so as to avoid data security breaches. The hosted computer environment within a data center must be explicitly managed, but most of the management is conducted in an automated fashion, thus saving hiring and energy costs. Data centers can be managed remotely and may not even house actual employees.
Functions of data center management include upgrading hardware and software/operating systems, managing data distribution and storage, backup regimes, emergency planning and some technical support.
There are many DCIM tools and suites of solutions on the market, and as with any acquisition, you need to look at each critically and choose the one that best meets your specific needs.
A cross connection is most often a layer 1 or physical layer connection between two networks. Data center providers typically segment cross connections by type of cabling used to make the connection - copper, coaxial or fiber. Cross connections are usually completed by the data center provider for a non-recurring (NRC) and a monthly recurring (MRC) charge.
Newer DCIM tools measure, monitor, automate, and optimize processes for energy efficiency. They can do things like:
"Critical" power or "IT load" often refers to the data center load that is consumed or is dedicated to IT equipment such as servers, storage equipment and communications switches and routers. Power for lighting or cooling the data center is excluded from "critical" power. It's important for an end user to understand their critical load as the data center - whether managed internally or outsourced - will be sized based on the current or expected amount of critical power.
There are two main categories of data center management software tools: monitoring/automation software and planning/implementation software.
The first deals with monitoring and automation of the IT room and facility power, environmental control, and security. It acts upon user-set thresholds by alarming, logging, or even controlling physical devices, and does things like verifying the data center is functioning as designed, and automating activities that optimize availability and efficiency.
The second category of software focuses on planning and implementation, where IT managers can typically have the greatest impact on total cost of ownership (TCO). It ensures efficient deployment of new equipment, organizes planning in order to facilitate changes in the data center, tracks assets, and simulates the impact of all kinds of what-if” scenarios.
The multi-tier data center model is dominated by HTTP-based applications in a multi-tier approach. The multi-tier approach includes web, application, and database tiers of servers. Today, most web-based applications are built as multi-tier applications. The multi-tier model uses software that runs as separate processes on the same machine using interprocess communication (IPC), or on different machines with communications over the network. Typically, the following three tiers are used:
Multi-tier server farms built with processes running on separate machines can provide improved resiliency and security. Resiliency is improved because a server can be taken out of service while the same function is still provided by another server belonging to the same application tier. Security is improved because an attacker can compromise a web server without gaining access to the application or database servers. Web and application servers can coexist on a common physical server; the database typically remains separate.
User interfaces — different packages offer different views, so choose those that would be most useful to you. Among those typically available:
With multiple virtual machines and applications running on any single host, the health and availability of each physical machine becomes that much more critical, and that’s where DCIM tools play a vital role in ensuring adequate power and cooling. The other consideration is the intensive and constantly changing power and cooling requirements of a virtual environment — dynamics loads simply can’t be responded to manually.
Cages and cabinets delineate the type of space that a colocation provider will convey to a customer in a retail colocation model. Cages are moveable walls on top of raised flooring to separate one customer's space from that of another. Cabinets, on the other hand, are typically lockable individual racks to house server, storage or communications equipment. Cages are typically for larger retail colocation customers, while cabinets can come in 1/3, half or full sizes. Both are used in shared room environments (other customers).
For years, IT managers have deployed their servers and IT equipment in hot and cold aisles. In such a scenario, the front side of two rows of equipment racks face each other and draw cool air into each rack's equipment intake. As such, the back side of two rows each expel hot air into the hot aisle. While this is an efficient concept, it may not go far enough for higher power loads. A solution to make the data center even more efficient is to deploy either hot or cold aisle containment. In cold aisle containment, the cold aisles are augmented to effectively "trap" cold air into the cold aisle. This allows the data center operator to increase air handling set points and more efficiently cool the intakes of the servers. On the other hand, hot aisle containment is a strategy to isolate the hot air exhaust found in the hot aisle. In both cases, the intent is to restrict mixing of significantly different air temperatures. Both solutions can be effective to lower PUE.