Fit for the Laptop to the NOC

Where does NeuralStar fit in your network? Anywhere from the laptop to the NOC.

NeuralStar is architected to be efficient, economical and have an appropriate footprint regardless of the size of the network being managed. In fact, rather than size, it is the level of complexity in the network that determines whether NeuralStar can be the right solution. After all, even the largest networks are relatively easy to manage if they are all SNMP and break no new application ground. On the other hand, even smaller networks that incorporate any of the following will find dramatic advantage from NeuralStar:

• heterogeneous devices in the same network, especially a mix of SNMP and non-SNMP-based devices
• the use of net-centered applications such as VoIP or videoconferencing.
• a need to collect different kinds of events—Fault, Performance, Security and more—on the same screens
• the need to simplify complex management data so that is understood and actionable by less-experienced administrators
• the network is managed as part of a tiered network structure, interlocking with other internal or external networks to share duties and resources

This last, for example, is a frequent characteristic of military networks that is becoming increasingly true in enterprises as well. Not necessarily large, tiered networks are characterized by a number of independently managed networks each with its own independent objective, at times being used and managed together to achieve a larger goal. For example, if there a natural disaster were to occur in a community, rescue workers from different cities, state militia and local police might be called to the scene. How will they connect and communicate with each other? Who will be responsible for managing networked resources in the field and back up the administrative lines?

Click to enlarge

This is exactly the kind of complexity NeuralStar was designed to manage. With its IER approach is that NeuralStar brings astronomically greater degrees of flexibility and decision-making control, which means network executives can implement the unique Concept of Operations (ConOps) they need and, in the case of tiered networks, multiple ConOps at once.

Exactly what is a ConOps? Different people have different names for it, but in a nutshell the ConOps is the plan expressing how you will manage your network based on business and mission goals. First, it involves setting policies such as Roles & Responsibilities, Procedures & Workflows and SLA & QoS Expectations. From there, it means defining the procedures and tools you will use to implement those policies, including Monitoring & Visualization, Coordination & Control, Filters & Thresholds, Event & Asset Definitions.

In simpler networks the ConOps is also simple, but complex networks require more flexibility and support from management tools than simply a flashing alert when a router went down. They require, contextual information, for example, and better control over how you define events, more background about alternate routing options and type of service, more options for representing collective data, and other things that only NeuralStar can provide.

NeuralStar’s architectural support for small, large and tiered networks includes:

• Node-based architecture enables the smallest footprint for different networks and host devices. Full NeuralStar functionality is available at all nodes and layers, regardless of size.
  
• Run multiple management protocols (SNMP and non-SNMP) on a single node, making it possible to manage anything from a few routers to a nationwide service network simultaneously from the same box using the same platform.
  
• Multi-tiered organizations can distribute network management capabilities across organizational levels with each network retaining autonomy over its areas of responsibility without compromising its subordinate roles in the enterprise strategy.
  
• Security and access rules are set independently to support top-down policies with horizontal control.
  
• Network integration can be accomplished either through web services or message layer, allowing managers of component networks to provide all data requested by managers without surrendering full system access. Uses web services and NeuralStar API to integrate with networks that are managed by other than NeuralStar solutions.
  
• Separate channels in the Message Layer for each type of data (alert, performance, configuration and inventory) allows NeuralStar to smoothly pass a Common Operational Picture (COP) and multiple business process views up, down and across the network hierarchy.
  
• Cross reference multiple naming conventions for a single object—including IP address, DNS name, sysName, engineering name and device aliases—enabling integration of data supplied by multiple management solutions.
  
• Device control responsibilities can be spread across multiple nodes, increasing the efficiencies of end-to-end service tasks in large or distributed networks.
  
• Traffic Throttling in the Messaging Layer maximizes bandwidth efficiency by directing local traffic to local nodes where information processing is preformed and reducing traffic at the control center.
  
• Employs a Publish/Subscribe messaging layer in which a single message traverses the system to reduce network traffic and improve efficiencies in scale.
• Correlations and Rules Processing (including thresholding and suppression) are executed at the Node level, providing autonomous network monitoring in mobile or dynamic networks while speeding overall network efficiencies in fixed network scenarios.

Learn More About NeuralStar:
Technology | Convergence | Visibility | Products | Tiering | Control