Network infrastructure up time and SLA’s are great but does that ensure value to the enterprise?
Network infrastructures these days are very robust and highly available due to fully meshed designs, HA protocols, etc, but does that automatically guarantee that the business is going realize a great user experience for business critical applications? Not really… IT infrastructures, as of late, are becoming more and more complex and the key, really, is visibility into the network. In addition to monitoring the fundamentals like router interfaces, load balancers, and firewalls, we should be looking at critical application performance and response times which translates into user experience and, in-turn, productivity for the enterprise.
So what do we do about it? Assuming that we have a solid working platform to start with, network visibility is the key. If a business critical application is slow, then why is it “slow”? Is it the network that is causing it? Is it the application tier or the backend? These are questions that are best answered by a good enterprise monitoring solution that has the ability to breakdown a transaction into its individual parts and do so in a way that is easy to view. A solution that presents information to the team that is actionable. Historically, this has been very difficult to put together and correlate, but now there are several solutions available that are able to make sense of all these disparate information streams and provide an enterprise with a clear picture. This insight will give a team the ability to be proactive in heading off potential problems in the enterprise.
So what is available? There are several top products that are out on the market, but let me warn you, they can be expensive. Some of these solutions have the ability to do session recording or packet capture so that you can go back and replay a session to determine what was happening at the time of the reported incident. This, in my opinion, is a very important tool to have in ones arsenal. This is also very valuable from a security perspective if you want to go back and potentially analyze a security incident.
Top Monitoring Tools:
- AppResponse Xpert – Opnet
- Riverbed Cascade
Once deciding on a solution that meets the requirements of the business, you can then breakdown the enterprise infrastructure into three parts:
- Data Collection
Data Collection – With most enterprise infrastructures, critical services are converged having voice, video, and data traversing the network, as well as managing key business critical applications. Some of these applications are very time sensitive, as in the case of voice services. The solution should be able to gather performance data from across the LAN and the WAN to paint an overall picture of the health of the network.
- Traffic – Users and systems on the network accessing what applications and protocols at any given time.
- Applications – What applications are active on the network. What is the performance of crucial applications and map that performance to the underlying infrastructure.
- Application Priority – Recognize if the network is delivering optimal performance for business critical applications.
- Packet – Capture and record all packets on the network for later analysis and investigation of performance degradation or security events.
Analytics – Automated analysis of the infrastructure and its services so that the IT group can make informed decisions to cut time while troubleshooting or develop strategies for capacity planning. Transform raw data into actionable information.
- Create baselines – What is normal behavior
- Trending – leverage historical data for future-state planning
- Monitor thresholds – Detect when links are being maxed out or CPU’s are reaching critical levels
- Detect Anomalies – Provide anomaly detection across multiple data sources and recognize suspicious or abnormal behavior.
- Alerting – Issue alerts for events that are deemed critical to the business
Reporting – Provide reporting that is relevant to the respective groups in the business
- Engineering – Detailed information reports to develop capacity planning strategies and drill down capabilities for real-views and troubleshooting
- Operations – High-level views about critical information and where their attention should be focused at a given time.
- Executives – “Big Picture” information service level tracking
There are many powerful tools on the market that can do the job, but it takes time and effort to implement these solutions properly. In addition, once implemented, they need to be maintained and updated to be effective. Many corporations neglect the care and feeding of these tools and they often fall by the wayside. Once this happens, IT is then flying blind and it is very difficult to justify the costs of re-implementing the solution.