As the network grows, the network equipment producers flourish, bringing many different exclusive products into the market. How to manage or operate so many equipment as the different vendors own diversified CLI and web interface to debug and configure. It’s time to put forward some new technologies, SDN vs. OpenFlow vs. OpenStack to tackle this problem.
SDN VS OpenFlow vs. OpenStack: What Are They?
SDN-Software Defined Network
Software-defined networking (SDN) technology is a new way to cloud computing.To improve network monitoring and performance, SDN is designed to enhance network management and promote programmatically network configuration efficiently. It centralizes network intelligence in one network component by decoupling the forwarding process of network packets (data plane) and the routing process (control plane). SDN is mainly composed by application layer which provides application and service, control layer responsible for unified management and control, and forwarding layer that offers hardware equipment like fiber switches, Gigabit Ethernet switches and routers to forward data. The following table illustrates the advantage of SDN against traditional network.
Software-defined Network vs. Traditional Network
Software-defined Network
Traditional Network
Forwarding and control separation
Forwarding and control coupling
Centralized control
Decentralized control
Programmable
Non-programmable
Open interface
Closed interface
OpenFlow: the Enabler of SDN
To turn the concept of SND into practical implementation, we need to put into place some protocols, among which OpenFlow is the most desirable one. So what is OpenFlow?
OpenFlow is a communications protocol that empowers a network switch or router to access the forwarding plane over the network. Also it can serve as a specification of the logical structure of the network switch functions. We know that each switch vendors may have their own proprietary interfaces and scripting languages, and this protocol enables them to work coordinately while avoid exposing their technology secret inside switches to the public.
OpenStack
OpenStack is an open source cloud computing management platform project that combines several major components to accomplish specific tasks. Its existence confronts the AWS of Amazon, as it allows all participators to access the source code and share some ideas, if they want to. It is convenient and reliable with strong compatibility and adaptability, gaining support from many vendors.
SDN vs. OpenFlow vs. OpenStack: What’s the Difference?
SDN vs. OpenFlow
SDN and OpenFlow are prone to be confused and misunderstood. Take a look at SDN vs. OpenFlow, the two are indeed interconnected. First of all, as an open protocol, OpenFlow underpins the various SDN controller solutions. The complete SDN solution is taking SDN controller as the core, backed by OpenFlow switches and NFV to offer bountiful SDN app for a new smart, dynamic, open, custom network.
OpenFlow vs. OpenStack
OpenFlow, since its release, has gained achievements in hardware and software support. CISCO, Juniper, Toroki and pronto have all launched network equipment like 10gbe switch, router, and wireless access point which support OpenFlow. In contrast, OpenStack covers many aspects like network, virtualization, operation system, and server. It is an ongoing cloud computing platform.
SDN vs. OpenStack
Network orchestration OpenStack copes with the component organization of a particular group of assets, from open source or closed implementations, thus we can say that it can be considered how a software-defined network is deployed. While SDN control serves like the commander of organizers and deals with maintaining consistent (as far as is feasible) policy across multiple groups of assets, so we deem it much like the "why."
Conclusion
SDN vs. OpenFlow vs. OpenStack, the three terms that are of far-reaching significance, attract more attention from the public. This article may provide you with some help to know them at the very first step. Till now, the networking technologies are still advancing, knowing what they are at present doesn’t mean the truly master of it. There is still plenty of room left to be explored.
We know that if any chain in the network fails, the operation may break down. Facing this problem, we’ve introduced the stackable switches and together with it is the concept of redundant links. When stacking switches, except the shortest link between switch and the main frame computer, we also prepare other links in case of the break down of the major link. The other links are redundant links between switches.
Introduction of Redundant Link
In order to maintain the stability of the network, composed of multiple switches, some backup connections are usually used to improve the robustness and stability of the network. The backup connection here is also called a backup link or a redundant link. Redundant links in a switches are accomplished through the use of multiple switches or multiple links between switches.
In an enterprise network, a link is redundant if its presence or absence does not affect the nature of the mechanism. That is, even if we remove that link, the mechanism will behave in the same manner.
Pros and Cons of Redundant Link
Pros
The redundancy in networks can improve its reliability. Our intention is that if one device fails, another can automatically take over. By adding a little bit of complexity, we try to reduce the probability that a failure in switch will take the whole network down. Spanning Tree Protocol,the redundancy protocols, can be implemented on any topology or mesh. The Cellular Redundancy provides alternative to running a physical line for redundancy. In addition, with Parallel Redundancy Protocol, we can achieve zero packet loss, “0ms” recovery. And it can be added to any existing network.
Cons
But you cannot have both complexity and reliability at the same time. The more complex something is, the harder it is to maintain, the greater the chance of human error, and the greater the chance of a software bug causing a new failure mode.
The switches between the backup links are often connected to each other to form a loop. The loops can be redundant to a certain extent. The redundant backup of the links can bring robustness, stability and reliability to the network. However, the backup link also causes loops in the network. The loop problem is the most serious problem faced by the backup link. The loop between the switches will cause new network problems: broadcast storm, loops and duplicate frames.
Tips
To make fully use of redundant links, we can minimize the complexity. Select two identical switches as the core switches. If you need gigabit Ethernet switch, for example, you can select two 10 gbe switches that run the same software and have the same connections. We can also introduce the Spanning Tree Protocol (STP) which was developed as a Layer 2 loop-avoidance mechanism for redundant links in a switched network. With STP, there will be only one logical path between all destinations on the network and redundant links that could cause a loop are intentionally blocked.
Conclusion
Redundant links are useful to a great extent. That’s why so many people now choose stackable switches rather than standalone ones to maintain the efficient network operation. Stackable switches are now our star products and focal point. We would like to introduce our high quality fiber switch to every people in need of reliable network performance.
عضو شوید
عضویت سریع
آمار مطالب
آمار بازدید
