RU186862U1 - Subscriber network device with virtualized network functions - Google Patents

Abstract

The utility model relates to data transmission systems, in particular to systems using subscriber network device virtualization (vCPE).

1. A subscriber network device with virtualized network functions, comprising a microcontroller, read-only memory, random access memory, connected to the microcontroller bus, information input ports connected by the address and data bus to the microcontroller, characterized in that a Wi-Fi unit is additionally introduced into it , an indication unit connected by a bus to the microcontroller, a RAID controller, connected by an address and data bus to the microcontroller, and with a read-only memory device as part of the HDD / SSD a high-speed packet data processing module with a non-blocking high-speed FPGA-based switching matrix, the first input / output of which is connected to the microcontroller by a bus; Ethernet transceiver modules, the first inputs / outputs of which are connected by bus to the second inputs / outputs of the high-speed packet data processing module with a non-blocking high-speed FPGA-based switching matrix; information input / output ports, the first inputs of which are connected to the second inputs / outputs of the Ethernet transceiver modules, and the second inputs / outputs are the inputs / outputs of the device.

2. The device according to p. 1, characterized in that all its elements are made using digital technology.

Description

The utility model relates to data transmission systems, in particular to systems using subscriber network device virtualization (vCPE).

Telecommunications operators are looking for new ways to generate revenue streams, more flexible organization of service delivery, reducing infrastructure maintenance costs and management automation. The answer to these needs is the concept of NFV (Network Function Virtualization) - virtualization of network functions. Solution developers are trying to offer operators new business cases that would make the use of technology cost-effective. Among the potentially attractive cases vCPE stands out (virtual Customer Premises Equipment or “virtualization of subscriber network devices”).

One of the main approaches to implementing vCPE is the Edge model, where all virtual functions are located on the client side on an inexpensive device (micro-server with a sufficient number of Ethernet ports), which is a local NFV infrastructure (NFVI) for a specific customer. When using the approach with creating a small virtual infrastructure on client devices, you do not need a huge pool of resources and costs appear only at the moment when the client orders the service.

The closest technical solution that meets the requirements of virtualization of subscriber network devices is the device described in the article "Evaluation of the effectiveness of network processor architectures" (Grishchenko V.I., Ladyzhensky Yu.V., Yunis M. The main directions of development of modern network processors D. Morgailov, Yu.V. Ladyzhensky, M. Yunis // Science and technology, DonNTU. - Donetsk, 2011. - (Series "Informatics, cybernetics and obucenuvlya tekhnika"). - No. 14 (188). - pp. 123-127 .) is a prototype.

This device contains a microcontroller, read-only memory, random-access memory, connected to the microcontroller by the bus, information input ports connected by the address and data bus to the microcontroller.

The purpose of the utility model is to provide hardware redundancy for the storage system, which is most critical for virtualization applications, when several virtual functions require sufficient storage system performance and high reliability.

This goal is achieved by the fact that in the device containing the microcontroller, read-only memory, random access memory, the address and data bus connected to the microcontroller, information input ports connected to the address and data bus with the microcontroller, an additional Wi-Fi unit, an indication unit, RAID controller connected by address and data bus with microcontroller and HDD / SSD; a high-speed packet data processing module with a non-blocking high-speed FPGA-based switching matrix, the first input / output of which is connected to the microcontroller by a bus; Ethernet transceiver modules, the first inputs / outputs of which are connected by bus to the second inputs / outputs of the high-speed packet data processing module with a non-blocking high-speed FPGA-based switching matrix; information input / output ports, the first inputs of which are connected to the second inputs / outputs of the Ethernet transceiver modules, and the second inputs / outputs are the inputs / outputs of the device.

Comparison with the prototype shows that the inventive device is characterized by the presence of new units and their connections between them. Thus, the claimed system meets the criterion of "novelty."

A comparison of the proposed solutions with other technical solutions shows that the listed elements used in the blocks are known, however, their introduction in this connection with other elements leads to the expansion of the functionality of the device. This confirms the conformity of the technical solution to the criterion of "significant differences".

In FIG. 1 shows a block diagram of the proposed device.

The device includes: microcontroller 1, high-speed packet data processing module with non-blocking high-speed FPGA-based switching matrix 2, read-only memory (ROM) 3, random access memory (RAM) 4, information input ports 5, Wi-Fi 6 unit , address and data bus 7, display unit 8, RAID controller 9, HHD / SSD 15 drives; ambient temperature sensor 10, temperature sensor of the cooling system 11, transceiver module Ethernet 12, 13; information input / output ports 14.

The proposed device operates as follows.

To connect remote offices, departments, etc., having their own local area network, using point-to-point, point-to-multipoint, and multipoint-to-multipoint services, it is necessary to install special equipment located on the customer’s side, but serviced centrally (for example, by a service provider or central office). In this case, the most modern approach is the introduction of the SDN / SD-WAN concept and support for NFV services implemented centrally (for example, in the operator’s data center) or distributed (both on vCPE, on the core network, and in the data center).

The introduction of NFV will allow operators and large corporate customers to transform the network infrastructure by eliminating the use of specialized network equipment in favor of fully software and virtualized solutions.

To use NFV technology, virtualization of subscriber equipment is required by using vCPE. When connecting a new client, in addition to installing a demarcation device that separates the client’s network and the operator’s network, the service provider often has the task of implementing additional functions - for example, to control and manage connections and traffic, solve client business problems, etc. So, many corporate customers require additional features such as a firewall, VPN support, and DDoS protection.

In order to implement such services today, you need to deliver, install, configure, and then also maintain the appropriate equipment on the client side. NFV helps to solve this problem in a much more efficient way, namely due to virtualization of network functions in software applications that can be run both on ordinary servers and on virtual machines running on these servers. In this case, the operator can, using only one network interface device to differentiate traffic, “place” all other functions, such as a firewall, on its territory, and to simplify the network and manage it, in particular for building service chains, implement SDN.

The main purpose of the proposed device is the construction of the SD-WAN architecture network by telecom operators and for corporate networks. For this, vCPE devices are used that are installed on the customer’s network side and allow implementing the SDN approach to network management and service implementation according to the NFV ideology.

To implement a subscriber network device with virtualized network functions (vCPE), a well-known structure for building network processors (prototype) with a central processing unit (CPU) 1, an address and data bus 7 connected to information input ports 5, and buses with read-only memory (ROM) was taken 3 and random access memory (RAM) 4. CPU 1 is also connected via a bus to a non-blocking high-speed FPGA-based switching matrix 2. The use of FPGA 2 chips provides the necessary flexibility in the development of hardware logic brabotki network traffic, as well as the possibility of implementing additional functionality when changing competitive environment without modifying the device architectures.

The inputs of the input ports 5 (Fig. 1) are connected to the sensors of the ambient temperature 10 and the temperature of the cooling system vCPE 11, which allows you to control the temperature mode of the device through CPU 1. In addition, the address and data bus 7 is connected to a Wi-Fi unit 6, which provides wireless monitoring of the device status, with an indication unit 8. The operation of this part of the device is controlled by CPU 1.

Ethernet Transceiver Modules (PHY) 12, 13 is an integrated circuit designed to perform the functions of the physical layer of the OSI network model.

PHYs allow other link-level chips, called MACs, to connect to a physical transmission medium, such as optical fiber or copper cable. The standard PHY microchip includes modules of the physical coding sublayer (PCS, Physical Coding Sublayer) and the transmission medium sublevel (PMD, Physical Medium Dependent). The physical coding sublevel module performs the functions of encoding and decoding the transmitted and received data stream. The purpose of coding is to simplify the process of recovering data stream by the receiver.

RAID controller operation 9.

Hardware RAID controller 9 is mounted in the subscriber switch. Module 9 of the proposed subscriber network device with virtualized network functions (vCPE) works as an independent microcomputer, as it has its own computing processor and memory. Due to this, the coordination of arrays occurs with minimal delays and without unnecessary calls to the central processor vCPE. This implementation gives a greater performance increase compared with integrated solutions when creating arrays from SSD drives or when creating high-level RAID arrays from more than six traditional disks. In addition, hardware RAID controllers are compatible with professional high-speed hard drives with SAS interface and can have additional means for increasing fault tolerance: their own cache memory on the board, the ability to work on battery power to save the contents of the cache memory in the event of a power failure and hardware correction errors on the fly. Hardware controllers, unlike software and integrated, support much more RAID schemes, some even provide the ability to use non-standard levels, and also allow you to build complex hybrid arrays from different RAID levels with the simultaneous use of hard and solid state drives.

Thus, the proposed subscriber network device with virtualized network functions (vCPE) will provide higher performance subsystem data storage (ROM) and guaranteed fault tolerance.

Claims (2)

1. A subscriber network device with virtualized network functions, comprising a microcontroller, read-only memory, random access memory, connected to the microcontroller bus, information input ports connected by the address and data bus to the microcontroller, characterized in that a Wi-Fi unit is additionally introduced into it , an indication unit connected by a bus to the microcontroller, a RAID controller, connected by an address and data bus to the microcontroller, and with a read-only memory device as part of the HDD / SSD a high-speed packet data processing module with a non-blocking high-speed FPGA-based switching matrix, the first input / output of which is connected to the microcontroller by a bus; Ethernet transceiver modules, the first inputs / outputs of which are connected by bus to the second inputs / outputs of the high-speed packet data processing module with a non-blocking high-speed FPGA-based switching matrix; information input / output ports, the first inputs of which are connected to the second inputs / outputs of the Ethernet transceiver modules, and the second inputs / outputs are the inputs / outputs of the device. 2. The device according to p. 1, characterized in that all its elements are made using digital technology.

Images