Hybrid network and IT infrastructures are becoming more complex every year, particularly with the deployment of software-defined networks (SDN), the adoption of cloud services, and the speed with which companies spin up new resources.
These technologies provide much greater flexibility and even simplicity to the overall business, but the engineers in the field are left with a daunting task. They must manage the convergence of these technologies while providing predictable and reliable end-to-end connectivity and quality of experience.
Organizations are looking for a single source of truth that unifies performance data across networks, applications, systems, and various technologies and protocols. But what does it take to deliver an effective unified monitoring platform?
Here are the top five monitoring essentials that I believe every IT professional should have at their disposal today. Take a look.
1. Speed at scale: An explosion of machine data persists in enterprise and service provider markets. Trends such as the virtual data center (which is quickly evolving to become the software-defined data center) and the Internet of Things, for example, exacerbate the issue. If network performance software can't scale to process and analyze massive, growing data sets instantly with the fewest possible resources, then all other capabilities become meaningless.
2. Data agnosticism: The network performance market is moving away from SNMP (Simple Network Management Protocol) and other standard protocols as a source of performance data. Flow, logs, APIs, and nonstandard data from proprietary business applications, network probes, and element management systems will take precedence.
Software that ingests and reports on any time series data, structured or unstructured, is the best way to go. It is important to be able to normalize and report on the data, so that uniform analytics can be run on it independent from its source.
3. Easy data in/out: The ability to easily ingest or export performance data will become increasingly important to the network performance market. Today, software that provides many ways to aid customers in getting data in/out of the platform for further analysis or manipulation will win over products that cannot provide this feature.
4. Reduced overhead: Organizations will continue to search for the broadest capabilities with the least possible investment. Network performance software that drives rapid time to value is essential and will ultimately result in ease of deployment and the reduction of administrative effort for IT professionals.
5. Expanded community: The human element of any performance tool is just as important as its technical capability. Community networking is a key component, and community forums, user support, and advice lead to a better understanding of your infrastructure performance.
I believe in network and infrastructure performance monitoring platforms that support a customer's need to spin up new services faster than before and exploit technologies that work for their business. The monitoring platform must prove infrastructure performance, but it also make recommendations on how to improve it continuously.
Does your network and infrastructure monitoring tool support the above functions? What other essentials do you think will be important in the future?
Bob Friday, Chief AI Officer for Juniper Networks, explains how the advanced technology is transforming operations.
Contact center leaders from 8x8, Awaken Intelligence, and 360insight discuss the importance of agent experience.
AI may force enterprises to rewire parts of their data centers so they are fully optimized to run such workloads. The question is do you use Ethernet or InfiniBand?
