In the discussion of options for a WiFi network billing system, one problem often arises: the V7 Authentication & Billing system itself has bandwidth control, DPI identification and strategy. Why buy a separate NE-80 current gateway?
Let's get both things straight.V7 is a software-forming Authentication & Billing operating platform, it runs on the server. It has a core function of authentication, set-up, billing, user administration, reporting, finance. It can do bandwidth control, identify private proxy, and send out strategies, but these capabilities are available as an attachment to NATSHEL_AUTH_BILLING, running on a universal server, with processing capacity limited by servers CPU and web cards.NE-80 is the hardware-forming layer flow control gatewayIt's a special box with links to the web export location, designed to provide deep package recognition and flow control in a mega-megae traffic environment. It can be 20G vomiting, 8 million co-connected, more than 3600 applications identified, 98% online.——These figures are the specifications of hardware boxes and are not easily tracked by software platforms on normal servers.
When the V7 is enough, when it's needed to add NE-80? The test is not complicated. If you have hundreds of people to one or two thousand, export bandwidth is in a gigabytes, and the main demand is certification and fees, and application recognition accuracy is low, then the V7 software platform with an ordinary export gateway will be enough, without hardware control.
But if your scene is...Campus, large campuses, hotel clusters, operator-level districtsThe export bandwidth is a mega-scale, 10,000 online users, P2P and video traffic, requiring precise restrictions on certain types of applications rather than simple speed limits. The software platform alone works. At this point, NE-80 sets of application layer recognition and flow control in the export location, V7 continues with NATSHEL_AUTH_BILLING, which combines: V7 tells NE-80"Who's the user, what kind of dinner package is this? How much bandwidth should I give you?"NE-80 is actually implemented in export locations"This user is watching video, downloading, running P2P, following a policy limit."。
The advantage of this division of labour is thatSeparation of performance and dutiesV7 server does not have to carry the depth recognition pressure of a megaflow. It only handles authentication and billing, thousands of people are also certified as steady; NE-80 hardware box is dedicated to doing what it does well.——The two devices are more robust than allowing a universal server to be used as an authentication server and as a mega-mode control.
NE-80 supports the bridge and route mode on deployment. The network mode is a web line that is smart, with minimal changes to the current network; it can be used directly as an export gateway for multiple loads. It also supports 8-way bridges, double-television, hardware sideways.——The hardware bypass is a key design: if the box itself breaks power or dies, the flow will go by automatically and the network will not be disrupted by a current control malfunction. This reliability is difficult for a software platform to run on a universal server.
The most confusing thing in the actual project is"DPI Capacity"Who's it to?The two DPIs are not the same thing: V7 DPI is designed to identify."Is this account shared and has a private router?"The results of its findings are used to bill and account for Anti-Sharing; the DPI of NE-80 is designed to identify"Is this traffic video or downloading or web page?"The results of the diagnosis are used for bandwidth control and application speed limits. One is a flow-oriented, judging purpose and configuration. Both devices do not conflict, but they need to know what each sees and not weigh on strategy.
There is also a real problem:How to trade off when the budget is limitedNot every project has to go to NE-80. If the customer exports bandwidth itself, and the number of users is small, hard plus a hardware flow box is wasteful. It's more practical to run on V7 platforms first, and then meet them."Some users run out of exits, application recognition is not good enough, software platforms are not holding up."When these particular issues are taken into account, add NE-80 to the export position. This step-by-step construction is more economical than starting with hardware piles. V7 and NE-80 are originally co-operatives, not necessarily buy together.
There is another scene to mention:Movement control of multiple export operators. If the park pulls both telecommunications and links more than two export routes, NE-80 can strategically channel different applications to different lines."Telecoms are on the telecommunication export, the Internet is on the connection."This multi-export route movement, which is more demanding on the software platform itself, and hardware gateways are more mature. V7 remains responsible for Authentication & Billing in this kind of a jump, leaving the export path to NE-80, with a clear division of labour.
This is a capability boundary: the performance parameters of NE-80 (20G throughput, 8 million connections, 98% recognition) are hardware specifications and how much can be achieved when actually deployed. It also needs to be field tested in conjunction with network pegging, commo models, application combinations."You'll be able to carry a trillions of money in this box."The specific performance is discussed in conjunction with the hardware specifications of the project and field pressure, which are clearly required by the knowledge base rule Kari.
Looking back, V7 and NE-80 are not options but a combination of software platforms plus hardware gateways. There is enough for the small-scale V7 platform; when one gestationary export, thousands online, requires precision application identification, it controls traffic in an export chain NE-80, and V7 continues to do NATSHEL_AUTH_BILLING, all divisions."We all have one."The phrase is biased, depending on its export bandwidth and user level, before deciding whether to add this hardware box.