What is Cloud Computing?
I know that today the buzz technology of the day is Cloud Computing, but is this really a new technology or just a new term for an old type of tech that fell out of favor. To understand this premise we need to look at what Cloud Computing promises both the user and management.
From Wikipedia we have this definition of Cloud Computing:
Cloud computing is Internet ("cloud") based development and use of computer technology ("computing"). From a technical point of view this generally involves the provision of dynamically scalable and often virtualised resources as a service over the Internet.[1][2] From a conceptual point of view, cloud computing refers to a paradigm shift in computing whereby computing resource and underlying technical infrastructure are abstracted away from the user. Users need not have knowledge of, expertise in, or control over the technology infrastructure in the "cloud" that supports them.[3] Cloud computing services often provide common business applications online that are accessed from a web browser, while the software and data are stored on the servers.
Very cool and interesting, but is it really different from the old Client/Server architecture from the good old days of mainframes and minicomputers? Here is the definition of Client-server again from Wikipedia:
Client-server computing or networking is a distributed application architecture that partitions tasks or work loads between service providers (servers) and service requesters, called clients.[1] Often clients and servers operate over a computer network on separate hardware. A server machine is a high-performance host that is running one or more server programs which share its resources with clients. A client does not share any of its resources, but requests a server's content or service function. Clients therefore initiate communication sessions with servers which await (listen to) incoming requests.
The only real difference other than updated terminology is that Cloud computing has moved Client-Server out to the internet. This is not a bad thing, it has some major advantages to the basic Client-server design, but is it really a new technology? The short answer is no, it is an old concept which has been given a new life by adapting for use over the internet and it is no longer confined to the realm of your LAN.
Technology Cycles vs. Business Cycles
After working in IT for many different businesses I have seen a reoccurring cycle or theme as to how best to serve users of Managed Information Services. While every tech likes to see his company explore the newest technology the reality is that the driving factor for every business in dealing with IT is money. “What is it going to cost?” is always asked and in today’s depressed economy that is an even more pressing issue. Most companies today buy or lease equipment on a three year cycle. With revenues falling the accounting department is looking for ways to stretch that cycle out to four or even five years, more if the economy stays down. I believe this is why the Client/Server server architecture is returning to popularity now with the new marketing campaign and rebranding as Cloud Computing.
A Client/Server architecture allows for the utilization of older less powerful computers with newer software as the processing load is kept mainly in the server. Over the last several years most companies have employed a File/Server architecture where most of the computing power was used at the client side and the server was mostly for Data storage. File/Server Architecture became prevalent as existing servers became older and new Client/Server type servers were cost prohibitive while desktop computers had fallen in price and the Return on Investment(RoI) to move to a File/Server architecture was sufficient to make the change for most companies. Today the winds of change have again shifted and now cost favor returning to a Client/Server architecture.
In conclusion if you are looking to predict the request of management just remember that the accountants have the most pull in any business, or at least they should, and as such they are your best source for help in building your case for a new technology to present to management. RoI will make or break your presentation in a down economy so make sure that you can show how any new tech you are asking for will save money for your boss. If you can demonstrate reduced cost by changing to a new technology your boss is much more likely to give you the green light to start designing or planning a new project.
Thursday, October 29, 2009
Tuesday, September 1, 2009
Understanding the Players of IT, a guide for Management
I know this is a little out of the typical strand of educating IT newbies, but I was asked today what a Network Eng. should know verses a System Admin and that got me to thinking about a couple of different things that I really wish Management would learn.
Now management will always ask more of their employees than they should, unless it is a governmental agency, but when it gets down too it what does a company need to cover their IT needs? I am going to go through and build out the perfect IT team and you see if you agree.
First, most Accountants try to work out the number of people in IT by how many people they support. This is a stupid thing accountants do, we just have to forgive them and move on so part of this discussion will work to justify the body count so that your CFO will sign off on it. Second, most IT managers truly have no clue as to what most of their people are doing at any given time of day, so I will try to include realistic activities that each worked would be tasked with performing on a daily basis. Third, most IT shops are overly weighted to one skill set which really makes it difficult to run efficiently. I will address each of these as I describe the positions. Now like any good team the players have to be able to perform in more than just their assigned position. Now, I would never take a goalie from a soccer team and ask them to suddenly start playing forward, however they could even if they weren't the best at it. The same is true of your IT employees, your Network Eng. should be able to support an end user if needed in a smaller shop, but it should never be one of his primary duties. On the other hand I could take a Guard from the football team and move them over to Tackle and expect them to perform the job with no problem and they should be able to handle the change and still be as good as they were at Guard. The same is true of your Network Eng. and your System Admin. in smaller shops. If you have a very limited number for your head count then some of your players really need to be well versed in multiple fields. Even in the larger shops there needs to be some over lap in knowledge simply for troubleshooting issues. I will be working from an IT shop which is budgeted for five or six employees, because we all know it doesn't matter how many you need, just what they will let you have. Well enough preaching lets get into the jobs:
Network Engineer:
The Network Eng. is you backbone, he will keep your remote sites up and the web pages flowing. He is not the guy most of your users will know real well as he is typically stuck in the basement or closet with all of the network equipment. While his primary responsibility is to keep the data flowing he should still be able to work on most Servers and be a fairly competent Server Admin when the network is calm. Every Network Eng I have ever known cringes at being called a programmer, but they should know the basics of scripting as well as some HTML, but don't depend on him to build complex Database Tables. Always remember not to over utilize this player as he should be the first responder to any outage and will be the employee with the most time spent at work after hours.
System Administrator:
The Sys Admin is your server guy, it is this player who will keep the servers working so users can access the files and programs they need to be productive. Most of your users will know his name and phone number but couldn't pick him out in the break room. The Sys Admin will spend most of his time working to keep the servers safe and secure with patches and updates, but he should also be able to deal with some of the basic networking needs. A Sys Admin should be able to configure a port on a switch for use with either a server or a user and he should also know what ports are needed for connectivity even if he doesn't know the command line interface for the firewall. Additionally, your Sys Admin needs to know not only how to install the Applications on the servers, but they should be fairly comfortable administering them and doing the basic as well as a few advanced functions with in them.
Your Sys Admin and Network Eng. should be very close. Any changers either of these two make could effect the others work and they need to communicate with each other constantly.
System Analyst/Programmer:
Not every shop has a Programmer type, but he is included here as a place holder for that guy you need to build a webpage or customize an app or even your Database Administrator. This is the job I never wanted to get stuck with. This is the worker who even when you know what he is doing, you don't really know what he is doing. Many shops have replaced the programmer with standard apps and outsourced support contracts, but most companies’ really still need one. A good System Analyst will be your go to employee for any issues with user interfaces or Database anomalies which need to be ferreted out before one of your customers end up with 1000 dohickies for $10 each instead of 10 thingies for $1000 each. A System Analyst should be as good of a Sys Admin as he is a Programmer, but he won't know nearly as much on the networking side. He does still need to know the ports for his apps and he will need to interface with the Network Eng. almost as often as the Sys Admin. does.
Many companies either drop the System Analyst and add an extra Sys Admin or they go to an in house development team. Which ever way you go make sure your player in this position knows his job duties and they are not included under the "other duties as assigned" catch all.
Help Desk Administrator:
This is the voice of IT to the users. He is often abused, and always under appreciated. No one ever calls the Help Desk to chat, they call because something doesn't work. Now while the Help Desk Admin will be out at users desk, this is the person who should be the go to guy for the users to resolve their problems or getting them to someone who can. This should also be the person who directs the work flow for any problems or trouble tickets. If you are lucky enough to have more than one Desktop Techniction then this player would be their coach making sure that the trouble tickets were divided out and taken care of. While this listing has the Help Desk Admin as the person who answers the help desk phone this would not be a Teir 1 support person. If you do have enough of a budget to have a Help Desk Tech for Tier 1 support then this would really be more of your Lead Tech, a Tier 2.5 support if you like while the players above are your Tier 3.
Desktop Support Technician:
Here finally we get to the face of IT. This is the poor guy who is sent out to the lions den to be eaten alive. Well, it's true, the life of the average Desktop support tech is short, if they are good at tech work they will get promoted as quickly as possible, if they aren’t any good they will end up making someone mad and get fired. All of the above players started out here, or if they didn't they should have. If you find someone to fill this job who is good at interacting with people and a good tech, you better pay him well. This is the guy who will have to interface not only with your users but every other player above. To be successful he will need to be able to talk to users with out talking down to them, yet he will also need to be able to translate the user’s words into something that the Admins and Programmers can understand. Additionally, a good Desktop Support Tech will be able to take that user who is mad and by the time his is done fixing the problem the user is not only happy, but now knows how to avoid the problem altogether. Desktop Support Techs are a dime a dozen, but good ones are worth their weight in trouble tickets.
Help Desk Technician:
Also known as Phone Support. This is Tier 1 support. Most IT shops do not have anyone dedicated to this job and instead have the phones set up to a hunt group so that everyone’s phone rings when someone calls into the Help Desk. If you do decide to fill this spot on your team then make sure this player is well versed in phone educate and has a good grasp of what kind of issues they might see. Since this is considered an entry level job it is wise to have your more experienced team members build documentation which will help with trouble shooting issues. These should be built in a question and answer format with the user's answers leading to the next question. Often when these scripts are built management forgets to include a point to hand off the call to the next level of support or if that is included it is not specified who is best suited to handle the call. This player for team IT is about the only one who can not step in to help out with other positions on the team, but they should be encouraged to help out and learn as they are the best source of new talent to promote as openings in the IT line up occur.
The above players are just a guide and you may find that you need additional team members or players who cover more ground than I have described. That is a common theme with in IT as there are always a large number of ways to do the same thing and none are inherently wrong. It is like Routing, there are people who stand by EIGRP and others who believe OSPF is best, yet in most instances RIP would work just as well.
Now management will always ask more of their employees than they should, unless it is a governmental agency, but when it gets down too it what does a company need to cover their IT needs? I am going to go through and build out the perfect IT team and you see if you agree.
First, most Accountants try to work out the number of people in IT by how many people they support. This is a stupid thing accountants do, we just have to forgive them and move on so part of this discussion will work to justify the body count so that your CFO will sign off on it. Second, most IT managers truly have no clue as to what most of their people are doing at any given time of day, so I will try to include realistic activities that each worked would be tasked with performing on a daily basis. Third, most IT shops are overly weighted to one skill set which really makes it difficult to run efficiently. I will address each of these as I describe the positions. Now like any good team the players have to be able to perform in more than just their assigned position. Now, I would never take a goalie from a soccer team and ask them to suddenly start playing forward, however they could even if they weren't the best at it. The same is true of your IT employees, your Network Eng. should be able to support an end user if needed in a smaller shop, but it should never be one of his primary duties. On the other hand I could take a Guard from the football team and move them over to Tackle and expect them to perform the job with no problem and they should be able to handle the change and still be as good as they were at Guard. The same is true of your Network Eng. and your System Admin. in smaller shops. If you have a very limited number for your head count then some of your players really need to be well versed in multiple fields. Even in the larger shops there needs to be some over lap in knowledge simply for troubleshooting issues. I will be working from an IT shop which is budgeted for five or six employees, because we all know it doesn't matter how many you need, just what they will let you have. Well enough preaching lets get into the jobs:
Network Engineer:
The Network Eng. is you backbone, he will keep your remote sites up and the web pages flowing. He is not the guy most of your users will know real well as he is typically stuck in the basement or closet with all of the network equipment. While his primary responsibility is to keep the data flowing he should still be able to work on most Servers and be a fairly competent Server Admin when the network is calm. Every Network Eng I have ever known cringes at being called a programmer, but they should know the basics of scripting as well as some HTML, but don't depend on him to build complex Database Tables. Always remember not to over utilize this player as he should be the first responder to any outage and will be the employee with the most time spent at work after hours.
System Administrator:
The Sys Admin is your server guy, it is this player who will keep the servers working so users can access the files and programs they need to be productive. Most of your users will know his name and phone number but couldn't pick him out in the break room. The Sys Admin will spend most of his time working to keep the servers safe and secure with patches and updates, but he should also be able to deal with some of the basic networking needs. A Sys Admin should be able to configure a port on a switch for use with either a server or a user and he should also know what ports are needed for connectivity even if he doesn't know the command line interface for the firewall. Additionally, your Sys Admin needs to know not only how to install the Applications on the servers, but they should be fairly comfortable administering them and doing the basic as well as a few advanced functions with in them.
Your Sys Admin and Network Eng. should be very close. Any changers either of these two make could effect the others work and they need to communicate with each other constantly.
System Analyst/Programmer:
Not every shop has a Programmer type, but he is included here as a place holder for that guy you need to build a webpage or customize an app or even your Database Administrator. This is the job I never wanted to get stuck with. This is the worker who even when you know what he is doing, you don't really know what he is doing. Many shops have replaced the programmer with standard apps and outsourced support contracts, but most companies’ really still need one. A good System Analyst will be your go to employee for any issues with user interfaces or Database anomalies which need to be ferreted out before one of your customers end up with 1000 dohickies for $10 each instead of 10 thingies for $1000 each. A System Analyst should be as good of a Sys Admin as he is a Programmer, but he won't know nearly as much on the networking side. He does still need to know the ports for his apps and he will need to interface with the Network Eng. almost as often as the Sys Admin. does.
Many companies either drop the System Analyst and add an extra Sys Admin or they go to an in house development team. Which ever way you go make sure your player in this position knows his job duties and they are not included under the "other duties as assigned" catch all.
Help Desk Administrator:
This is the voice of IT to the users. He is often abused, and always under appreciated. No one ever calls the Help Desk to chat, they call because something doesn't work. Now while the Help Desk Admin will be out at users desk, this is the person who should be the go to guy for the users to resolve their problems or getting them to someone who can. This should also be the person who directs the work flow for any problems or trouble tickets. If you are lucky enough to have more than one Desktop Techniction then this player would be their coach making sure that the trouble tickets were divided out and taken care of. While this listing has the Help Desk Admin as the person who answers the help desk phone this would not be a Teir 1 support person. If you do have enough of a budget to have a Help Desk Tech for Tier 1 support then this would really be more of your Lead Tech, a Tier 2.5 support if you like while the players above are your Tier 3.
Desktop Support Technician:
Here finally we get to the face of IT. This is the poor guy who is sent out to the lions den to be eaten alive. Well, it's true, the life of the average Desktop support tech is short, if they are good at tech work they will get promoted as quickly as possible, if they aren’t any good they will end up making someone mad and get fired. All of the above players started out here, or if they didn't they should have. If you find someone to fill this job who is good at interacting with people and a good tech, you better pay him well. This is the guy who will have to interface not only with your users but every other player above. To be successful he will need to be able to talk to users with out talking down to them, yet he will also need to be able to translate the user’s words into something that the Admins and Programmers can understand. Additionally, a good Desktop Support Tech will be able to take that user who is mad and by the time his is done fixing the problem the user is not only happy, but now knows how to avoid the problem altogether. Desktop Support Techs are a dime a dozen, but good ones are worth their weight in trouble tickets.
Help Desk Technician:
Also known as Phone Support. This is Tier 1 support. Most IT shops do not have anyone dedicated to this job and instead have the phones set up to a hunt group so that everyone’s phone rings when someone calls into the Help Desk. If you do decide to fill this spot on your team then make sure this player is well versed in phone educate and has a good grasp of what kind of issues they might see. Since this is considered an entry level job it is wise to have your more experienced team members build documentation which will help with trouble shooting issues. These should be built in a question and answer format with the user's answers leading to the next question. Often when these scripts are built management forgets to include a point to hand off the call to the next level of support or if that is included it is not specified who is best suited to handle the call. This player for team IT is about the only one who can not step in to help out with other positions on the team, but they should be encouraged to help out and learn as they are the best source of new talent to promote as openings in the IT line up occur.
The above players are just a guide and you may find that you need additional team members or players who cover more ground than I have described. That is a common theme with in IT as there are always a large number of ways to do the same thing and none are inherently wrong. It is like Routing, there are people who stand by EIGRP and others who believe OSPF is best, yet in most instances RIP would work just as well.
Thursday, August 27, 2009
Complaints!
I thought we were past WordStar2000! What the hell is up with the formatting on this blog? WYSIWYG too complicated for them to figure out?
Getting started
Getting started is a lot harder than I thought. I have started looking at some of the other blogs out in this wide world of 1s and 0s and the first thing I notice is there are a lot of 0s out in the world. It seems that there are many people with Blogs who might best serve the human race by not reproducing. There is a old (well in IT anything over a year is old) saying "There are 10 types of people in the world, those who understand binary and those who don't." I know that probably started as a math joke, but it belongs to IT now so live with it. The sad thing is there are a lot of people in IT that don't get it.
I have found that one of the hardest things to teach someone is really one of the simplist concepts in computers and that is the concept of +1. Now, I know that there are plenty of people who will want to cuss me out, but it is true. If you can not add in Binary then you can't subnet IP addresses. If you can't work with Subnets, then you can't network successfully. I have worked for a lot of different companies and over the years I have found many examples of subnets not correctly defined.
Everyone who has set up a computer knows the following subnet: 255.255.255.0 That is the classic Class C subnet, nothing special, nothing hard. If you have ever set up a serial link or a link to a service provider they have used 255.255.255.252 , so most people can recognise that one as well. Most people haven't used them but still know the following: 255.0.0.0(Class A) and 255.255.0.0(Class B). But when was the last time you set up a corporate office with 255.255.252.0? Can you work out what your Network address, Broadcast Address and First and Last available IPs are? Well, if you would like to not only know how to work out this needed information but to also learn what it all means then please dear reader continue on.
First of all you must understand what 255 and 0 represent. 255=11111111 and 0=00000000 Thats not hard at all, and you get to deal with four of those groups called Octets. Second, you need to understand how to the subnet relates to the IP address. To put it simply If the subnet is a 0 it doesn't matter, if it is a 1 it has to match. Dosen't really make sense does it? I'll draw a picture:
IP Address: 10.10.10.10=00001010.00001010.00001010.00001010
Subnet: 255.255.252.0 =11111111.11111111.11111100.00000000
As you can see the first two octets are all 1s and must match perfectly to the network address. The last octet is all 0s and doesn't matter what you have in there. This makes the third octet the important part, so lets pull out just the third octet and see what we have there.
0 0 0 0 1 0 1 0 I spaced out the digits to make
1 1 1 1 1 1 0 0 it line up better.
Now at this point you know that those two 0s in the subnet are allowed to be anything they want to be as long as the first 8 stay the same. So how do you change this into an IP Address that you can type into a Windows TCP/IP properties window? Well you learn the idea of +1.
Now what do I mean by +1. Well in school you learned that 1+1=2, well that would be great if 2 was a number that we could use, but we can't. We are stuck in binary, it is either on or off. Works like a light switch in gandma's house and is represented as either 0 or 1. Now look at this count to ten in decmal and binary:
Decmal: Binary:
0 0 Always start with 0
1 1 Pretty easy here, nothing weird happening
2 10 What happened we are already at 10?
3 11
4 100 Are you worried yet? Don't be it will make sense in a minute
5 101
6 110
7 111
8 1000
9 1001
10 1010 Can you see the pattern now?
With only two digits we have to use a lot more places to show the same value. This is why IP addresses are restricted to a top value of 255. Looking back at our IP and subnet above we can see that only those last two digits can change which gives us the following choices:
0 0 0 0 1 0 0 0 = 8
0 0 0 0 1 0 0 1 = 9
0 0 0 0 1 0 1 0 = 10
0 0 0 0 1 0 1 1 = 11
So we now see that there are a total of four values that can fit with in a subnet of 252(11111100) for those keeping tract of the Binary. So now we see that with in that network scope we can have 8, 9, 10 and 11 as values in the third octet.
Now to look at the subnet again:
255.255.252.0 = 11111111.11111111.11111100.00000000
As discussed previously if the digit is a 1 is must remain consistant, and if it is a 0 it doesn't matter. There are two exceptions, the Network address and the Broadcast address. With in the subnet scope neither the Network nor the Broadcast addresses can be used. all other addresses are usable. So, how do we know what these addresses are? Well the network address is all 0s when the subnet is all 0s, while the Broadcast is all 1s when the subnet is all 0s. I'll demonstrate:
Network : 0 0 0 0 1 0 1 0. 0 0 0 0 1 0 1 0. 0 0 0 0 1 0 0 0. 0 0 0 0 0 0 0 0
Subnet : 1 1 1 1 1 1 1 1. 1 1 1 1 1 1 1 1. 1 1 1 1 1 1 0 0. 0 0 0 0 0 0 0 0
This shows that our network address will be 10.10.8.0. I have bolded the significant digits. The Broadcast address is done with the same places, but with 1s:
Network : 0 0 0 0 1 0 1 0. 0 0 0 0 1 0 1 0. 0 0 0 0 1 0 1 1. 1 1 1 1 1 1 1 1
Subnet : 1 1 1 1 1 1 1 1. 1 1 1 1 1 1 1 1. 1 1 1 1 1 1 0 0. 0 0 0 0 0 0 0 0
From this we see that the broadcast address is 10.10.11.255. This means that we now know that 10.10.8.0 to 10.10.11.255 is our scope and 10.10.8.1 to 10.10.11.254 is the usable range. That's a lot of Addresses.
Now one thing which you will need to remember is that idea of +1. You have seen 1+1=10, but what happens when you have an octet of 11111111? We need to look at a pair of octets which we will have one of all 0s and the other of all 1s, and then we add 1.
00000000.11111111
+1
00000001.00000000
Just think of it like your odomator, only you have just 0s and 1s. And that is all of the Binary I will force feed you today. Honestly to need to know more would make you a math major not a computer science major.
Many years ago I was shown a chart that helps with subnetting. I mean this thing makes an real differance when you are working on these caculations. But there is a down side, if you don't understand what has already been discussed you will have a heck of a time remembering what to do to make the chart work. Now that I have offered up that little disclaimer here it is:
128 192 224 240 248 252 254 255
128 64 32 16 8 4 2 1
To remember how to build this chart remember that the top row is you subnet and the bottom row is the size of the range. Second notice that 255+1=256, 128+128=256, and that is true of every pair. This is because there are 256 possible combination of 0's and 1's in an octet. So to start using this from the chart you know that with a 240 subnet there are 16 addresses, but remember you have to subtract two, one is the network and one is the broadcast and neither are usable. Another useful rule from this chart is that each of the numbers in the bottom row is evenly divisable by the smaller numbers(i.e. 16/4=4, 64/32=2). This means that with a .240 subnet you know that .16 is a network address and so is .32, .64, and .128. This gives you a starting point to work from when you get an address like 10.10.10.96 subnet 255.255.255.240. If you need to find out if 10.10.10.99 is in the same network you need to beable to get up to that range with out having to do 16, 32, 48, 64, 80, 96, and 112. We see from what I just worked out that .96 is the network address which would explain why a computer assigned that IP and Subnet wouldn't talk on the network, but we could have skipped most of that math by starting at 64 since we know it is a network address of .240 subnet.
Now what information is needed most of the time when you are subnetting an IP range? You need to know the Network address, Broadcast Address, # of Avaible Addresses, and most people what to know the first and last avaible addresses. Here is our IP and subnet:
65.204.122.36
255.255.255.252
This is for your serial interface on your new Cable modem and you need to program it for use, but something is not working right. Lets work out the subnet information and see if we can figure out what is wrong. A .252 subnet gives us a range of four with two usable. 32 is closest to 36 and I am betting that you can already see that 36 is a network address in a .252 subnet. So what do we now know about this subnet:
Subnet 255.255.255.252
Network 65.204.122.36
First 65.204.122.37
Last 65.204.122.38
Broadcast 65.204.122.39
Next Network 65.204.122.40
We can now see that the addresses for the interfaces should have been .37 and .38. This is the process to deal with subnet issues most of the time.
When someone is needing to subnet a range of addresses into smaller ranges it is because they have a limited range and need to use it in several different places. If we are given the Class C range of 205.55.146.0, but we need to use this range in multiple locations or in several different VLANs then we need to beable to further divide these addresses.
Ok, so here is a situation which could happen to you as a network admin. The office manager wants to divide users by department and segment the network using differnet VLANs. Here is the list of departments with the number of users:
HR 6
Sales 22
Mangement 5
Finance 13
IT 3
I know the first thing you notice is that IT has 3 people, well that is not realistic, but I didn't want him to be lonely. Now, as you can see none of these groups will need to take up a full Class C scope, nor will all of the groups together take up a full Class C, but we should always plan for growth. It really sucks when you have to start adding additional IP ranges to an area because someone forgot to plan ahead. So the other information you will need to know is how many people can we expect to add? How many printers are needed? And are we going to have IP phones? Here we can see that we need a lot more addresses than what the numbers above indicated. So here is what we find out after we ask some questions of the different departments:
Department: Employees: Printers: IP Phones: Future Growth:
HR 6 4 10 4
Sales 22 6 35 10
Mangement 5 5 5 1
Finance 13 8 18 5
IT 3 1 5 2
Now we can divide this up in many differnet ways. Here you just have to make some educated guesses. First thing you have to decide is do you want the printers and phones on their own VLANs? Next, do you think their growth estamates are close? I'm going to always recomend putting phones in their own VLAN, and the same with printers if you can. So here is what we will need with the current numbers we have gathered:
HR 10
Sales 32
Mangement 6
Finance 18
IT 5
Printers 14
Phones 73
Now that we have a feel for the number of addresses needed in each scope you will be building we can get to work on figuring out what scopes will meet our needs. If you will go back up and look at our chart we can see that the number of addresses needed for the phones is the largest and will only fit with in a 128 subnet. So with out needing to worrie about it further we know that we will have to use the following for the phones:
Network 205.55.146.0
Subnet 255.255.255.128
First Availble 205.55.146.1
Last Availble 205.55.146.126
Broadcast 205.55.146.127
At this point we know that we have used half of our IP addresses. So we look at the rest of the groups.
HR 10
Sales 32
Mangement 6
Finance 18
IT 5
Printers 14
Sales looks easy with 32, we can use a .224 and have 32 addresses, but wait. We don't have 32 addresses, we have 30, always remember to take out two to account for the Network and Broadcast addresses. So now we have to take Sales up to a 192 subnet.
Network 205.55.146.128
Subnet 255.255.255.192
First Availble 205.55.146.129
Last Availble 205.55.146.190
Broadcast 205.55.146.191
It looks like we are getting a little tight on IPs. Lets look at the rest and I will add in what subnet we wouild like to be using:
HR 10 .240
Mangement 6 .240
Finance 18 .224
IT 5 .240
Printers 14 .224
Now, I know that Management could fit into a .248, but there would be no room to add any more devices. This also is true of the printers fitting into a .240. They will fit, but you have no space if someone decides to add another. So can we make all of these fit. Start with the largest scopes first and work it out:
Finance
Network 205.55.146.192
Subnet 255.255.255.224
Printers
Network 205.55.146.224
Subnet 255.255.255.224
Opps, we are out of addresses. We can contact our service provider and get more, but how many do we need? Lets work it out as the next scope:
HR
Network 205.55.147.0
Subnet 255.255.255.240
Management
Network 205.55.147.16
Subnet 255.255.255.240
IT
Network 205.55.147.32
Subnet 255.255.255.240
We will need to use up to 205.55.147.47, but an ISP is not going to give us three .240 subnets, they will give us one .192 subnet though and now we have a little room to grow.
Now you can see how to divide out these IP addresses to fit the needs of your company. I hope it helped you out at least a little.
If you want a little practice copy the following and work them out and email them to me at jsorrells@gmail.com and I will try to get them back to you as quickly as possible.
Practice #1
235.11.4.119
255.255.255.64
Practice #2
10.20.30.40
255.255.224.0
Practice #3
192.168.55.0
255.255.255.254
Practice #4
172.16.20.255
255.255.248.0
Practice #5
172.20.20.11
255.248.0.0
Practice #6
10.255.255.28
255.255.255.252
Practice #7
12.4.4.240
255.255.255.192
Practice #8
4.4.4.190
255.255.224.0
Practice #9
10.44.12.188
255.255.255.0
Practice #10
10.100.20.200
255.192.0.0
Bonus Questions:
Which of the following IPs are usable given the assigned subnet?
Which IPs are included in the 224.128.60.0 Network?
224.128.61.255
255.255.252.o
224.128.62.255
255.255.252.o
224.128.61.0
255.255.252.o
224.128.64.0
255.255.252.o
224.128.60.254
255.255.252.o
I have found that one of the hardest things to teach someone is really one of the simplist concepts in computers and that is the concept of +1. Now, I know that there are plenty of people who will want to cuss me out, but it is true. If you can not add in Binary then you can't subnet IP addresses. If you can't work with Subnets, then you can't network successfully. I have worked for a lot of different companies and over the years I have found many examples of subnets not correctly defined.
Everyone who has set up a computer knows the following subnet: 255.255.255.0 That is the classic Class C subnet, nothing special, nothing hard. If you have ever set up a serial link or a link to a service provider they have used 255.255.255.252 , so most people can recognise that one as well. Most people haven't used them but still know the following: 255.0.0.0(Class A) and 255.255.0.0(Class B). But when was the last time you set up a corporate office with 255.255.252.0? Can you work out what your Network address, Broadcast Address and First and Last available IPs are? Well, if you would like to not only know how to work out this needed information but to also learn what it all means then please dear reader continue on.
First of all you must understand what 255 and 0 represent. 255=11111111 and 0=00000000 Thats not hard at all, and you get to deal with four of those groups called Octets. Second, you need to understand how to the subnet relates to the IP address. To put it simply If the subnet is a 0 it doesn't matter, if it is a 1 it has to match. Dosen't really make sense does it? I'll draw a picture:
IP Address: 10.10.10.10=00001010.00001010.00001010.00001010
Subnet: 255.255.252.0 =11111111.11111111.11111100.00000000
As you can see the first two octets are all 1s and must match perfectly to the network address. The last octet is all 0s and doesn't matter what you have in there. This makes the third octet the important part, so lets pull out just the third octet and see what we have there.
0 0 0 0 1 0 1 0 I spaced out the digits to make
1 1 1 1 1 1 0 0 it line up better.
Now at this point you know that those two 0s in the subnet are allowed to be anything they want to be as long as the first 8 stay the same. So how do you change this into an IP Address that you can type into a Windows TCP/IP properties window? Well you learn the idea of +1.
Now what do I mean by +1. Well in school you learned that 1+1=2, well that would be great if 2 was a number that we could use, but we can't. We are stuck in binary, it is either on or off. Works like a light switch in gandma's house and is represented as either 0 or 1. Now look at this count to ten in decmal and binary:
Decmal: Binary:
0 0 Always start with 0
1 1 Pretty easy here, nothing weird happening
2 10 What happened we are already at 10?
3 11
4 100 Are you worried yet? Don't be it will make sense in a minute
5 101
6 110
7 111
8 1000
9 1001
10 1010 Can you see the pattern now?
With only two digits we have to use a lot more places to show the same value. This is why IP addresses are restricted to a top value of 255. Looking back at our IP and subnet above we can see that only those last two digits can change which gives us the following choices:
0 0 0 0 1 0 0 0 = 8
0 0 0 0 1 0 0 1 = 9
0 0 0 0 1 0 1 0 = 10
0 0 0 0 1 0 1 1 = 11
So we now see that there are a total of four values that can fit with in a subnet of 252(11111100) for those keeping tract of the Binary. So now we see that with in that network scope we can have 8, 9, 10 and 11 as values in the third octet.
Now to look at the subnet again:
255.255.252.0 = 11111111.11111111.11111100.00000000
As discussed previously if the digit is a 1 is must remain consistant, and if it is a 0 it doesn't matter. There are two exceptions, the Network address and the Broadcast address. With in the subnet scope neither the Network nor the Broadcast addresses can be used. all other addresses are usable. So, how do we know what these addresses are? Well the network address is all 0s when the subnet is all 0s, while the Broadcast is all 1s when the subnet is all 0s. I'll demonstrate:
Network : 0 0 0 0 1 0 1 0. 0 0 0 0 1 0 1 0. 0 0 0 0 1 0 0 0. 0 0 0 0 0 0 0 0
Subnet : 1 1 1 1 1 1 1 1. 1 1 1 1 1 1 1 1. 1 1 1 1 1 1 0 0. 0 0 0 0 0 0 0 0
This shows that our network address will be 10.10.8.0. I have bolded the significant digits. The Broadcast address is done with the same places, but with 1s:
Network : 0 0 0 0 1 0 1 0. 0 0 0 0 1 0 1 0. 0 0 0 0 1 0 1 1. 1 1 1 1 1 1 1 1
Subnet : 1 1 1 1 1 1 1 1. 1 1 1 1 1 1 1 1. 1 1 1 1 1 1 0 0. 0 0 0 0 0 0 0 0
From this we see that the broadcast address is 10.10.11.255. This means that we now know that 10.10.8.0 to 10.10.11.255 is our scope and 10.10.8.1 to 10.10.11.254 is the usable range. That's a lot of Addresses.
Now one thing which you will need to remember is that idea of +1. You have seen 1+1=10, but what happens when you have an octet of 11111111? We need to look at a pair of octets which we will have one of all 0s and the other of all 1s, and then we add 1.
00000000.11111111
+1
00000001.00000000
Just think of it like your odomator, only you have just 0s and 1s. And that is all of the Binary I will force feed you today. Honestly to need to know more would make you a math major not a computer science major.
Many years ago I was shown a chart that helps with subnetting. I mean this thing makes an real differance when you are working on these caculations. But there is a down side, if you don't understand what has already been discussed you will have a heck of a time remembering what to do to make the chart work. Now that I have offered up that little disclaimer here it is:
128 192 224 240 248 252 254 255
128 64 32 16 8 4 2 1
To remember how to build this chart remember that the top row is you subnet and the bottom row is the size of the range. Second notice that 255+1=256, 128+128=256, and that is true of every pair. This is because there are 256 possible combination of 0's and 1's in an octet. So to start using this from the chart you know that with a 240 subnet there are 16 addresses, but remember you have to subtract two, one is the network and one is the broadcast and neither are usable. Another useful rule from this chart is that each of the numbers in the bottom row is evenly divisable by the smaller numbers(i.e. 16/4=4, 64/32=2). This means that with a .240 subnet you know that .16 is a network address and so is .32, .64, and .128. This gives you a starting point to work from when you get an address like 10.10.10.96 subnet 255.255.255.240. If you need to find out if 10.10.10.99 is in the same network you need to beable to get up to that range with out having to do 16, 32, 48, 64, 80, 96, and 112. We see from what I just worked out that .96 is the network address which would explain why a computer assigned that IP and Subnet wouldn't talk on the network, but we could have skipped most of that math by starting at 64 since we know it is a network address of .240 subnet.
Now what information is needed most of the time when you are subnetting an IP range? You need to know the Network address, Broadcast Address, # of Avaible Addresses, and most people what to know the first and last avaible addresses. Here is our IP and subnet:
65.204.122.36
255.255.255.252
This is for your serial interface on your new Cable modem and you need to program it for use, but something is not working right. Lets work out the subnet information and see if we can figure out what is wrong. A .252 subnet gives us a range of four with two usable. 32 is closest to 36 and I am betting that you can already see that 36 is a network address in a .252 subnet. So what do we now know about this subnet:
Subnet 255.255.255.252
Network 65.204.122.36
First 65.204.122.37
Last 65.204.122.38
Broadcast 65.204.122.39
Next Network 65.204.122.40
We can now see that the addresses for the interfaces should have been .37 and .38. This is the process to deal with subnet issues most of the time.
When someone is needing to subnet a range of addresses into smaller ranges it is because they have a limited range and need to use it in several different places. If we are given the Class C range of 205.55.146.0, but we need to use this range in multiple locations or in several different VLANs then we need to beable to further divide these addresses.
Ok, so here is a situation which could happen to you as a network admin. The office manager wants to divide users by department and segment the network using differnet VLANs. Here is the list of departments with the number of users:
HR 6
Sales 22
Mangement 5
Finance 13
IT 3
I know the first thing you notice is that IT has 3 people, well that is not realistic, but I didn't want him to be lonely. Now, as you can see none of these groups will need to take up a full Class C scope, nor will all of the groups together take up a full Class C, but we should always plan for growth. It really sucks when you have to start adding additional IP ranges to an area because someone forgot to plan ahead. So the other information you will need to know is how many people can we expect to add? How many printers are needed? And are we going to have IP phones? Here we can see that we need a lot more addresses than what the numbers above indicated. So here is what we find out after we ask some questions of the different departments:
Department: Employees: Printers: IP Phones: Future Growth:
HR 6 4 10 4
Sales 22 6 35 10
Mangement 5 5 5 1
Finance 13 8 18 5
IT 3 1 5 2
Now we can divide this up in many differnet ways. Here you just have to make some educated guesses. First thing you have to decide is do you want the printers and phones on their own VLANs? Next, do you think their growth estamates are close? I'm going to always recomend putting phones in their own VLAN, and the same with printers if you can. So here is what we will need with the current numbers we have gathered:
HR 10
Sales 32
Mangement 6
Finance 18
IT 5
Printers 14
Phones 73
Now that we have a feel for the number of addresses needed in each scope you will be building we can get to work on figuring out what scopes will meet our needs. If you will go back up and look at our chart we can see that the number of addresses needed for the phones is the largest and will only fit with in a 128 subnet. So with out needing to worrie about it further we know that we will have to use the following for the phones:
Network 205.55.146.0
Subnet 255.255.255.128
First Availble 205.55.146.1
Last Availble 205.55.146.126
Broadcast 205.55.146.127
At this point we know that we have used half of our IP addresses. So we look at the rest of the groups.
HR 10
Sales 32
Mangement 6
Finance 18
IT 5
Printers 14
Sales looks easy with 32, we can use a .224 and have 32 addresses, but wait. We don't have 32 addresses, we have 30, always remember to take out two to account for the Network and Broadcast addresses. So now we have to take Sales up to a 192 subnet.
Network 205.55.146.128
Subnet 255.255.255.192
First Availble 205.55.146.129
Last Availble 205.55.146.190
Broadcast 205.55.146.191
It looks like we are getting a little tight on IPs. Lets look at the rest and I will add in what subnet we wouild like to be using:
HR 10 .240
Mangement 6 .240
Finance 18 .224
IT 5 .240
Printers 14 .224
Now, I know that Management could fit into a .248, but there would be no room to add any more devices. This also is true of the printers fitting into a .240. They will fit, but you have no space if someone decides to add another. So can we make all of these fit. Start with the largest scopes first and work it out:
Finance
Network 205.55.146.192
Subnet 255.255.255.224
Printers
Network 205.55.146.224
Subnet 255.255.255.224
Opps, we are out of addresses. We can contact our service provider and get more, but how many do we need? Lets work it out as the next scope:
HR
Network 205.55.147.0
Subnet 255.255.255.240
Management
Network 205.55.147.16
Subnet 255.255.255.240
IT
Network 205.55.147.32
Subnet 255.255.255.240
We will need to use up to 205.55.147.47, but an ISP is not going to give us three .240 subnets, they will give us one .192 subnet though and now we have a little room to grow.
Now you can see how to divide out these IP addresses to fit the needs of your company. I hope it helped you out at least a little.
If you want a little practice copy the following and work them out and email them to me at jsorrells@gmail.com and I will try to get them back to you as quickly as possible.
Practice #1
235.11.4.119
255.255.255.64
Practice #2
10.20.30.40
255.255.224.0
Practice #3
192.168.55.0
255.255.255.254
Practice #4
172.16.20.255
255.255.248.0
Practice #5
172.20.20.11
255.248.0.0
Practice #6
10.255.255.28
255.255.255.252
Practice #7
12.4.4.240
255.255.255.192
Practice #8
4.4.4.190
255.255.224.0
Practice #9
10.44.12.188
255.255.255.0
Practice #10
10.100.20.200
255.192.0.0
Bonus Questions:
Which of the following IPs are usable given the assigned subnet?
Which IPs are included in the 224.128.60.0 Network?
224.128.61.255
255.255.252.o
224.128.62.255
255.255.252.o
224.128.61.0
255.255.252.o
224.128.64.0
255.255.252.o
224.128.60.254
255.255.252.o
Wednesday, August 26, 2009
Getting started, Random Thoughts on Wednesday Afternoon
I need to write a book, but about what.
What am I going to put up here?
I will need to remember that I used my real name and avoid posting anything that could cost me my job.
I hope that I can modify the page layout and personalize it to fit my taste.
I will have to remember that I do not have a spell checker and will have to act accordingly.
Oooo, there is a spell checker up on the tool bar. Thank goodness.
I just love watching people complain about a political party, I guess they really don't have any clue that neither side is interested in what they want.
Every Politiction in the world is interested in getting re-elected, and they will do what ever it takes to get it. The only differance is who they pander too.
I emailed Rick Perry's office yesterday asking to be appointed to Key Bailey Huchersons Senate seat she just vacated. You think he will call me back? Yea, I'm not holding my breath.
I'll get back after work this evening and post something worthy of the readers I don't have yet.
What am I going to put up here?
I will need to remember that I used my real name and avoid posting anything that could cost me my job.
I hope that I can modify the page layout and personalize it to fit my taste.
I will have to remember that I do not have a spell checker and will have to act accordingly.
Oooo, there is a spell checker up on the tool bar. Thank goodness.
I just love watching people complain about a political party, I guess they really don't have any clue that neither side is interested in what they want.
Every Politiction in the world is interested in getting re-elected, and they will do what ever it takes to get it. The only differance is who they pander too.
I emailed Rick Perry's office yesterday asking to be appointed to Key Bailey Huchersons Senate seat she just vacated. You think he will call me back? Yea, I'm not holding my breath.
I'll get back after work this evening and post something worthy of the readers I don't have yet.
Subscribe to:
Posts (Atom)
Posts
