Redefining Analytics Deployment
Many analytics solutions capture and analyze network traffic in order to detect performance problems or security threats on the network. Usually these are hardware-based appliances that tap into the network in order to access the traffic they analyze.
The Current Problem with Analytics Deployment
The process of deploying these analytics appliances in a large, global enterprise network is complex, expensive and lengthy. As a result, changing analytics capability or deploying new analytics can take months or even years.
- Because analytics appliances are expensive, a CAPEX budget must be raised. This can take six to twelve months, depending where the enterprise is in its budgeting cycle.
- Once budget is approved, the next step is usually an RFP process. This can take three months.
- A shortlist of possible solutions are typically evaluated in a “bake-off” process. This bake-off generally takes another three months.
- Once a solution has been chosen, the purchasing and deployment process starts. In large, geographically distributed networks this can take six to nine months.
- With the appliances purchased and deployed, the process of doing northbound integration, and integration with other tools, starts. This can take another three to four months.
What’s the Cause of the Problem?
There are three main causes.
- The functionality of appliance-based analytics solutions is inextricably bound into the hardware. That means deploying new functionality requires a truck-roll to deploy new appliances.
- Analytics appliances are expensive. That automatically means a large capital investment and the associated lengthy CAPEX and procurement processes that entails.
- Deploying hardware into a complex enterprise network is not a trivial exercise. The process must be carefully project-managed. Installation and northbound integration has to be scheduled, and rack space must be made available.
Each of these things take time. Combined, they take forever.
How Deploying Analytics on an Analytics Platform Changes the Game
Packet capture functionality is an essential requirement for many analytics solutions. Currently each appliance has its own packet capture functionality built-in.
Which means you pay for packet capture many times over – part of the reason why deploying multiple analytics appliances is so expensive.
Our analytics platform provides two core functions. It captures and records Network History down to the packet level. And it provides hosting for analytics applications. Hosted applications can access this captured packet data in real-time or replay recorded Network History to do back-in-time analysis (read The Anatomy of an Analytics Platform for further detail).
This means that you only have to do packet capture once. Once the packets have been captured and recorded, they can be made available to all the analytics tools that need access to them – either by hosting them on the platform, or using the API to search for, and retrieve, recorded Network History. As someone once said “Endace takes the hard out of hardware”.
Deploying an analytics platform means just one hardware deployment. Once the platform has been deployed and the northbound integration is done, future deployments of analytics functions can be done in software, without hardware truck-rolls. Upgrades or swap-outs are just as easy.
Deploying analytics functions in software offers the flexibility, speed, agility and cost savings that virtualization has delivered in the datacenter. Evaluating and deploying analytics functions can be done in days, not months. You can change solutions as your needs change without ripping-and-replacing existing hardware.
Don’t you think it’s about time the rules of analytics deployment were redefined?
Yes I'd Like a Demo
How about a Demo?
No longer, do you need to rip-and-replace analytics appliances as your needs change. Book a demo to see how a single platform solution will give you more for your money and your network.
Find out just how fast and accurate your investigations could be.