Analyze The Nature And Volume

Published Date: 02 Nov 2017

Our goal is to understand why their network been built and what sort of users & applications they have been used.

To improve poor performance

To enable new applications

To apply more security to their system

Upgrade old equipment and gain more speed

Because of above problems they desire to upgrade their network system according their geographical scope. So we trying to analyze and decide what will be the best e commerce infrastructure for that company and plan according to their capacity.

We identified what will be our objectives will be.

To define the geographical scope of the system

To define application and users that will use the network

About Company Nature

When we analyzing their Head office at Colombo 07, we identified what sort of communication they are doing in day to day business.

Their Head office is a 4 storied building.

They have including 9 branches across the country

They have one LAN and one WAN.

They using Hubs for data transmitting through the Local Area Network.

Security is less according to their capacity of computers.

Routers are using for their Wide Area Network.

All other branches to gain information have to access the Head Office main server first.

They are using 4-pair Twisted Pair cabling.

They Have 5 Logistic Departments in their office

Account Department

Information Technology Department

Marketing Department

Human Resource Department

Financial Department

IT department can access all the computers of other Departments.

Other All Departments shouldn’t access to other departments without authorization of the manager of each department.

They are not using security types such as domain controller.

They have a lack security system which we identified and trying to propose a far better system than they had.

They using Windows XP and windows server 2008 in client and server computers.

Users have to wait half n hour for the print job to download.

Network Design

They didn’t use a special technique to design their network. We decide to approach to the Building Block Network Design according to their needs which we done earlier.

Building Block Network Design

C:\Users\Salinda Malalgoda\Desktop\img005.gif

At the above of this documentation we done the first stage of building block network design which is need analyzing and but some parts still we have to do.

As we know the first step in Needs Analyzing Break the network into three conceptual parts (based on their scope).

As we decide we trying to first break up their Local Area Network into Virtual Local Area Networks.

Virtual LANs (VLANs)

Virtual LANs (or VLANs) separate a Layer-2 switch into multiple broadcast domains. Each VLAN is its own individual broadcast domain (i.e. IP subnet). Individual ports or groups of ports can be assigned to a specific VLAN. Only ports belonging to the same VLAN can freely communicate; ports assigned to separate VLANs require a router to communicate. Broadcasts from one VLAN will never be sent out ports belonging to another VLAN. Please note: a Layer-2 switch that supports VLANs is not necessarily a Layer-3 switch. A Layer-3 switch, in addition to supporting VLANs, must also be capable of routing, and caching IP traffic flows. Layer-3 switches allow IP packets to be switched as opposed to routed, which reduces latency.

So we planned to set a VLAN to each department and then we can use a domain controller to restrict their access to other department’s computers. So Security part will be at the bottom of this documentation and we first stepped into decide what is the best e-commerce infrastructure to this company. So as we decide to propose 5 Virtual Local Area Network and combine them into one Local Area Network. Dedicated Servers can use for different purposes such as File Servers, Communication Servers, Print Servers by each department.

As we know Information Technology Department should be able to access all the computer in all department for many reasons.

To maintain Pc faults.

To maintain network characteristics etc.

So we proposed to combine all VLAN into 1 Local Area Network in the building. At the later stage in this documentation we can see what we proposed Devices, Security and Costing Assessments.

Wide Area Network

According to first step in Needs Analyzing we need a core layer which is consisting with Distribution layers in LAN. As we know the company has 10 branches and need to transfer data with and make some transactions with the head office and other branches. So we decide to every Service branch that wants to access other branches need to access through Head office Server and do the transaction. We already know our Internet Provider will be SLT Telecom and we communicate with them and discuss about their Data links which we need to make this WAN smoothly. So through this WAN any branch can access Head office server and they can communicate with any branches also.

Problems We Looking For

We concern about mainly what are the issues regarding to our design and we mainly looking into Bandwidth Vs Latency commonly say slow network.

We try to propose a network which gives "FASTER" internet connection that’s our main responsible. In this organization there are 10 branches so Wide Area Network Should be there. We think about a business that has a head office with a share point server and a branch office where users use that Head office’s server over the designed Wide Area Network. So our main concern is if enough users call into the helpdesk to complain about slow connection what do we do?, Then company and other people should be able to buy a bigger WAN link which is wrong thing to do which is what’s really going on. As I mentioned earlier there two things to measure when we bother about company’s network link.

Bandwidth – This is how much data can we transfer at once. For example, in one packet from the source to destination we measure this in KBPS, MBPS, or even in GBPS if the company has lot of money.

Latency – This is how much time takes to transfer one packet from its source to destination. An electron can only travel at 1 speed on a copper wire and increasing the bandwidth have no affect on this. Finally devices that process the data at various hops as an example routers, switches and firewalls, also add to latency. So, the further apart a client and server are, the longer transmission takes. Adding in more network devices, e.g. transmissions between different ISP’s, worsens this.

There is a way to handle when any data or file is transferred over the network between the client and the server. The file is broken up into packets and Headers and controls flags wrap each of those packets up to increase the amount of data. Here’s how it goes,

Server : Packet 1 comes

Client : Acknowledgment came

Server : Packet 2 comes

Client : Acknowledgment came

Server : Packet 3 comes

Client : Acknowledgment came

Server : Packet 4 comes

Client : Acknowledgment came

Server : Packet 5 comes

Client : Acknowledgment came

Server : Packet 6 comes

Client : Acknowledgment came

Server : Packet 7 comes

Client : Acknowledgment came

It goes until all of the packets that make up the entire file are transmitted. We identified that bandwidth also affects the time for transmission by reducing how much data we can put into a packet and note that the TCP stack in our Operating System can also limit this scenario. As we know we try to describe that bandwidth also causes problems when we try to put too many simultaneous conversation onto a pipe. We can monitor it by measuring link utilization.

Latency we can explain to the company as this. I it takes 1 millisecond to transmit a packet between the server and client then the above packets in whole file copy would take 14milliseconds. We thought that as we move the client to a remote location then latency goes up, perhaps to 100milliseconds. Now the file copy takes 100 times longer: 1400milliseconds, realistically, a file transfer requires exponentially more packets and an intercontinental latency measurement (use PING) might be 300 milliseconds or more!

We try to convert this problem to this company also and let’s go back to the above examples and see how this company will affected them

By throwing bandwidth at the share point server, we allow many more users in the branch offices in this company to have the same slow experience. Latency causes the packets that make up the file transmission to be slow and their some business opportunities will be in danger.

Solutions we looking for

When we looking into design of this network, there’s a few solutions we can look at:

One major thing we thought is to place the servers closer to the clients which their previous network system unable to did. For a "cloud computing" service provider, that’s possible to have service closer to the consumers. For a corporate, this might mean having servers in the branch office, something we want to steer clear of doing if at all possible to reduce costs and complexity.

Reduce the hops between the client and server by subscribing to major Internet Service Provider such as SLT Telecom. But there’s always going to be clients who many hop away which we cannot do anything to avoid it.

Some Advance Solutions

When a PC in the branch office access to a copy of file from the Head office server all security stuff will be involved with its transaction. The appliance at the Head Office server silently between the server and the WAN connection and it listens to the streams and using has algorithm try to break down the data into some blocks which we planned to stored on the appliance according to a set of predefined rules. When the data travels begin over the WAN connection to the branch office, the branch office appliance also listens to the new data and before caching them break in to blocks by using the hash algorithm. At this time no speed variance can be identified but the data stream continues to the client.

Think that another client in the same branch try to transfer this same file again then second client does the usual file lock and security stuff which are involve within it. Server believes talking to the client instead it talk to the appliance and make sure same data blocks will not be transmitted again. This will reduce the bandwidth utilization and also by removing the need to send data at all we can reduce the latency also. Other than our security techniques file transferring can be local only at the branch its means between the appliance and the client.

So the system will work by using blocks the optimization will come to party when files that haven’t been requested yet over the WAN before.

Talking About Appliance

We describing that appliances work at a TCP level and this means WAN optimization can imrove by several ways, e.g. Exchange, Oracle, SQL, Lotus Notes, tec. So the basic requirements are the data is not encrypted and not signed and the company also need to shut down SMB data signing in Group Policy because the appliances are in a way performing a man in the middle attack and we also consider that these appliances are very much expensive and we make sure they will not widespread.

Propose Next Generation TCP Stack

As we consider about problems in WAN we try to introduce new TCP stack to this company and to their network system. This TCP stack isn’t the complete WAN solution but it imrove some areas of the WAN. Microsoft included the new TCP stack in Vista, Windows 7 and in Windows server 2008 and windows server 2008 R2. This will reduce our latency and here is a best example according to what we explained earlier when transfer packets before waiting for an acknowledgment.

Server : Packet 1

Server : Packet 2

Server : Packet 3

Server : Packet 4

Server : Packet 5

Server : Packet 6

Server : Packet 7

Client : I want Packet 3 again

Server : Packet 3

Client : Acknowledgment

As we explained in earlier example to transfer packets it took 1400 millisecond if our latency was 100 milliseconds but our new TCP stack will do it in 900 milliseconds.

Windows 7 & windows server 2008 R2 better together

As we know they are using windows XP and windows server 2008 we want to replace their operating system with windows 7(ultimate & enterprise) and windows server 2008 R2. This will allow Win 7 clients to access a branch office cache of whole data which are stored on a windows server 2008 R2 server and the protocols being optimized are SMB (file sharing), Http and Https which is logically, BITS which are 2 architectures.

Let’s find out with an example,

Client : Clients in the branch office which have a peer to peer network and he wants to download a file from the Head office server for a business purposes. Then all security stuff is done as this process. The client broadcast based on the Local Area Network to see whether other clients also have the file cached and it uses the hash ID for the file which is obtained from the main server. The file is downloaded from the head office server if no other client no access to it. After all security stuff another client will come and try to download another file and again it broadcasts using that hash ID. When First client responds and the second client transfer file over the LAN which is not the WAN. This uses one broadcast domain and it is limited to a VLAN. And then another problem is branch office’s cached content disabling because their computers hibernated or shuttled down.

Hosted: As we mentioned earlier in the branch office we replace server with Windows server 2008 R2 and then branch office client communicate with the server by using uncast instead of broadcast bases p2p model. This allow multiple VLAN’s because network tidies up.

We want to make sure and explain them that Branch Cache only supports file based, not block level, caching and it also only caches when download and when uploading(saving) will be transferred over the WAN to the main(central) server and which is not optimized.

So finally we are planned to having the client (pc) in the branch office and roaming on the network and the server in the head office. We’ll always have latency when we transferring data between the remote client and the server even we described above solutions. What happens if closest client (user) logged in to the server may be central client can be access from anywhere it doesn’t matter from in a branch office, hotel, even at home but make sure data travels only over the WAN, not through the LAN.

As we know their users(clients) have to wait half n hour to print job to download and Windows server 2008 R2 solved that biggest problem which they have all printer drivers thanks to Easy Print. And also thanks to Microsoft XPS technology and make sure client runs Windows XP SP3 – 7. And Microsoft also added some applications like SSL interface which have ability to securely access those from TS gateway.

Windows server 2008 R2 called this as RDS (Remote Desktop Services) because VDI broker to access virtual desktops but we have risk in our hand to introduce to it because it have still its BETA version. By using that,

We can access familiar desktop applications.

We can reuse some administrative Systems.

Applications’ problems will reduce.

Helpdesk doesn’t need to change control to fix user issues.

So Our Main Goal is to give and describe and propose a better logical and physical design network system to the company which have more performance and less issues with it. And we try to use the latest technology design to implement a better network system and try to gain more & more performance than earlier System.