OSPF and BMG Usage in Networking



  • What is OSPF and BMG in networking?
    Where it is used and for Why it is important to use?


  • Service Provider

    They are routing protocols. They are the protocols that routers use to talk to each other to determine where routes exist. Not used in normal business networks until you get to incredible size.



  • @scottalanmiller This was my one of my interview question where i did not replied to him.



  • all the time i have this confusion, why during our academic studying and exams they give too much emphasize to this kind of stuff that are related specifically to ISPs and very big environment not normal businesses, because the majority of us work in normal enterprises and not specifically ISPs,


  • Service Provider

    @IT-ADMIN said:

    all the time i have this confusion, why during our academic studying and exams they give too much emphasize to this kind of stuff that are related specifically to ISPs and very big environment not normal businesses, because the majority of us work in normal enterprises and not specifically ISPs,

    No idea. Why is big networking such a focus at all? Systems administration is often ignored even though 99.99% of companies use it and everyone talks about network administration which only exists for 10% of companies!!



  • for example someone work in a normal business, does he need one day to set up routing protocols ?? or maybe i'm working right now in a small business therefor i cannot see the importance of routing protocol because i only have one router ??



  • I definitely felt this way with regards to a lot of the topics covered by the MCSE NT 4.0, 2000, 2003. 2008 seemed to remove a lot of focus ISDNs, T1s, etc.



  • @IT-ADMIN said:

    for example someone work in a normal business, does he need one day to set up routing protocols ?? or maybe i'm working right now in a small business therefor i cannot see the importance of routing protocol because i only have one router ??

    As Scott said, only about 10% of companies are large enough to worry about this. Though I suppose you don't have to be huge to need to worry about his. An associate of mine worked for a 250 person company and they considered their services so critical to keep online, they two ISPs able to handle the same IP range on both. (damn lost the term for that.)


  • Service Provider

    @Dashrender said:

    @IT-ADMIN said:

    for example someone work in a normal business, does he need one day to set up routing protocols ?? or maybe i'm working right now in a small business therefor i cannot see the importance of routing protocol because i only have one router ??

    As Scott said, only about 10% of companies are large enough to worry about this. Though I suppose you don't have to be huge to need to worry about his. An associate of mine worked for a 250 person company and they considered their services so critical to keep online, they two ISPs able to handle the same IP range on both. (damn lost the term for that.)

    It makes sense that a company of any size might have a unique need for that kind of failover, but that would be even more rare because you don't need routing for internal use failover. This means that they were hosting their own "critical" services. Something you don't normally do. One would question what set of circumstances made them think that they were so critical as to need to do that but not critical enough to have their externally facing systems in a datacenter?



  • @scottalanmiller said:

    @Dashrender said:

    @IT-ADMIN said:

    for example someone work in a normal business, does he need one day to set up routing protocols ?? or maybe i'm working right now in a small business therefor i cannot see the importance of routing protocol because i only have one router ??

    As Scott said, only about 10% of companies are large enough to worry about this. Though I suppose you don't have to be huge to need to worry about his. An associate of mine worked for a 250 person company and they considered their services so critical to keep online, they two ISPs able to handle the same IP range on both. (damn lost the term for that.)

    It makes sense that a company of any size might have a unique need for that kind of failover, but that would be even more rare because you don't need routing for internal use failover. This means that they were hosting their own "critical" services. Something you don't normally do. One would question what set of circumstances made them think that they were so critical as to need to do that but not critical enough to have their externally facing systems in a datacenter?

    Honestly, something they just probably never considered. They always did it in house, why move it out - but today we know why, but it still takes someone thinking of it before you even have the conversation.


  • Service Provider

    @Dashrender said:

    Honestly, something they just probably never considered. They always did it in house, why move it out - but today we know why, but it still takes someone thinking of it before you even have the conversation.

    Using a datacenter for hosting has been a standard since the 1990s and the idea that you never run critical apps in house since the early 2000s. It's a long time to have avoided conversations and not thought about best practices. At some point "status quo" in an organization isn't a good excuse because you can't have had the same people in place all that time without outside experience or exposure.


  • Service Provider

    @scottalanmiller said:

    Using a datacenter for hosting has been a standard since the 1990s and the idea that you never run critical apps in house since the early 2000s.

    In your opinion. Not the rest of the real SMB world.


  • Service Provider

    @JaredBusch said:

    @scottalanmiller said:

    Using a datacenter for hosting has been a standard since the 1990s and the idea that you never run critical apps in house since the early 2000s.

    In your opinion. Not the rest of the real SMB world.

    Best practices have been there. Even basic certs were requiring these concepts in the 1990s or CompTIA in 2000. Not just my opinion, industry opinion and best practices. Just because the SMB often eschews best practices doesn't mean that they aren't there.



  • @scottalanmiller said:

    @Dashrender said:

    Honestly, something they just probably never considered. They always did it in house, why move it out - but today we know why, but it still takes someone thinking of it before you even have the conversation.

    Using a datacenter for hosting has been a standard since the 1990s and the idea that you never run critical apps in house since the early 2000s. It's a long time to have avoided conversations and not thought about best practices. At some point "status quo" in an organization isn't a good excuse because you can't have had the same people in place all that time without outside experience or exposure.

    These are the Scott standards since those time frames. Perhaps even then large corporation standards, but definitely not the standard seen in use/practice at the SMB level. While I barely consider email critical (I know others do) there are still tons of SMBs that run it in house and only now are looking at things like Rackspace and O365.

    I may have been considered a best practice to move everything to a datacenter since the 1990's (I missed that memo though), but as we know, the filtering of information is sometimes, perhaps even often, slow to be passed along, let alone adopted.

    You have the benefit of being extremely logically minded and self motivated to do the best thing possible, most businesses, as we can see from many examples in SW, do not. As such, you discovered long ago that you could save money by hosting your infrastructure in a DC versus handling it onsite - taking into account everything from bandwidth, power, heating/cooling, environmental controls.

    And while I do believe all of that to be true from your numbers, It's still going to cost a company more money than self hosting. If I'm in a DC that means I need two ISP connections (the one in the DC - which probably comes with my rental space) and my ownsite - no doubling up there = more spending.


  • Service Provider

    @Dashrender said:

    These are the Scott standards since those time frames. Perhaps even then large corporation standards, but definitely not the standard seen in use/practice at the SMB level. While I barely consider email critical (I know others do) there are still tons of SMBs that run it in house and only now are looking at things like Rackspace and O365.

    These are things taught in the industry. If you have to resort to calling things "Scott standards" you should probably stop and think if you aren't looking for excuses. I was taught this stuff as very basic IT in the late 1990s. Ever since Spiceworks began people have been talking about this. In no way is this a "Scott thing." That SMBs do IT poorly doesn't imply that standards and best practices, common sense or good ideas aren't known - only that people are lazy, don't care or get "weird" and think that they know better than the industry.

    That lots of SMBs do it actually tells us that likely it is true as we know that SMBs very often do very poor IT decision making both on the technology and on the business side. This doesn't mean that it is wrong in absolutely every case, but we expect that anything that is a well known best practice will broadly not be followed in the SMB for a very, very long time, if ever. So that it is seen there tells us in no way that the people working there didn't know better or that the best practice didn't exist.

    Using "SMBs do bad things" as an excuse for SMBs to keep doing bad things is partially why they do it - it's become socially acceptable in technology circles to excuse bad decisions because a company is an SMB.


  • Service Provider

    @Dashrender said:

    I may have been considered a best practice to move everything to a datacenter since the 1990's (I missed that memo though), but as we know, the filtering of information is sometimes, perhaps even often, slow to be passed along, let alone adopted.

    Exactly. That they didn't care to follow best practices, pay attention, listen or whatever doesn't suggest that they didn't know or they practices didn't exist.


  • Service Provider

    @Dashrender said:

    And while I do believe all of that to be true from your numbers, It's still going to cost a company more money than self hosting.

    So you believe and don't believe at the same time? Why will it cost more? I know companies that have saved money doing it.


  • Service Provider

    @Dashrender said:

    If I'm in a DC that means I need two ISP connections (the one in the DC - which probably comes with my rental space) and my ownsite - no doubling up there = more spending.

    This isn't a logical point. Yes, "you" need more connections. That's like saying that every website you use requires you to have a connection for their server to the Internet. Yes, technically it does. Does using a website = more spending? No. So this doesn't either. You've taken a fact (that the servers need a network connection) and made a disconnected assumption that this will make it more costly. This simply isn't a direct correlation.


  • Service Provider

    @scottalanmiller said:

    These are things taught in the industry.

    Bullshit.
    I was never taught these things.
    I have been doing this truly profesisonally since graduating from ITT (waste of money) in 1993.
    Never in any of my cert courses or anything else was I ever "Trained" in these things you claim.
    I went back for a full degree in 2005, eventually getting a my Master's degree.
    None of this was ever taught.

    This is a reality in your world. Not the entire world.


  • Service Provider

    @JaredBusch said:

    @scottalanmiller said:

    These are things taught in the industry.

    Bullshit.
    I was never taught these things.
    I have been doing this truly profesisonally since graduating from ITT (waste of money) in 1993.
    Never in any of my cert courses or anything else was I ever "Trained" in these things you claim.
    I went back for a full degree in 2005, eventually getting a my Master's degree.
    None of this was ever taught.

    This is a reality in your world. Not the entire world.

    Likewise, you are taking your own experience and claiming that no one knows things because you don't know them. You aren't the end all of knowledge and experience.

    I'm sorry that your degrees and such missed stuff like this, but that's something we talk about all the time how that education process causes people to miss things, especially the basics.

    CompTIA has made this part of their entry level cert process in 2000. Microsoft was teaching it before then. College rarely teaches IT as a practice or best practices or even hires people that work in IT to do so, so I'm not sure why that's even mentioned.

    Just because you've missed things that the industry has been discussing doesn't mean that it isn't there. I'm taking the position that I have no special exposure, and I did not, and that the general learning processes which I did, which were less than you, covered this as obvious, basic and standard.

    And since then, that has been supported by years of discussions in the industry. Not that everyone follows best practices, but almost no one disputes what they are or that they exist.


  • Service Provider

    I'll admit, assuming that my experience isn't special holds people to a fairly high bar, in theory. But the alternative is worse, right?


  • Service Provider

    @scottalanmiller said:

    Likewise, you are taking your own experience and claiming that no one knows things because you don't know them. You aren't the end all of knowledge and experience.

    I am not claiming that no one knows. Then again, I am not claiming everyone knows either. You are.



  • @scottalanmiller said:

    @Dashrender said:

    I may have been considered a best practice to move everything to a datacenter since the 1990's (I missed that memo though), but as we know, the filtering of information is sometimes, perhaps even often, slow to be passed along, let alone adopted.

    Exactly. That they didn't care to follow best practices, pay attention, listen or whatever doesn't suggest that they didn't know or they practices didn't exist.

    I didn't know. I didn't learn this until the mid 2000's.


  • Service Provider

    @Dashrender said:

    @scottalanmiller said:

    @Dashrender said:

    I may have been considered a best practice to move everything to a datacenter since the 1990's (I missed that memo though), but as we know, the filtering of information is sometimes, perhaps even often, slow to be passed along, let alone adopted.

    Exactly. That they didn't care to follow best practices, pay attention, listen or whatever doesn't suggest that they didn't know or they practices didn't exist.

    I didn't know. I didn't learn this until the mid 2000's.

    That's still a full decade ago, right? About the amount of time that I stated originally. Maybe a small amount less, but not much. And were you an IT decision maker by that time?



  • To make these claims as standards, at least to me, means that the average IT person would have this knowledge, and hopefully be applying it. I say that this is not the case. That the average IT person didn't know this, and therefore wasn't following it.

    You'd an admitted mega reader - this alone puts you outside the norm. I'd be willing to wage that the average IT person reads one maybe two IT books a year at best, more likely they are keeping up only on the IT things that personally interest them.

    Where were you working when you learned that critical services should be moved to a DC and not locally hosted? what were the circumstances of that learning?

    I'm asking because I feel that they will be unusual circumstances compared to the experiences of most IT, especially SMB IT personal.


  • Service Provider

    @Dashrender said:

    To make these claims as standards, at least to me, means that the average IT person would have this knowledge, and hopefully be applying it. I say that this is not the case. That the average IT person didn't know this, and therefore wasn't following it.

    This is where I don't agree. Maybe not the average IT person if you include help desk and other non-decision making roles that aren't associated with this kind of decision. If you use that inclusion then I would agree that there isn't a single thing in IT that is "standard knowledge."

    But for people tasked with this kind of decision making, I am saying that it was extremely common knowledge by then. Maybe not every single person knew it, but average I'm quite confident did. Applying it, mostly, but not as commonly as know it.


  • Service Provider

    @Dashrender said:

    Where were you working when you learned that critical services should be moved to a DC and not locally hosted? what were the circumstances of that learning?

    Likely at a hotel doing the overnight desk shifts while working to get back into IT. I did most of my learning then when there was time before getting into IT (after coming out of software engineering.) I had a few years where I was managing restaurants and auditing hotels while switching career modes. That was the late 1990s. Definitely by 1999 I was talking to hospitals about the need to be in a datacenter for critical workloads, I remember the conversation (and the hospital losing their data during the meeting, which was awesome.)


  • Service Provider

    @Dashrender said:

    I'm asking because I feel that they will be unusual circumstances compared to the experiences of most IT, especially SMB IT personal.

    Unusual I don't think. I think most people had time where they learned about IT before working in the field. Not all, clearly some jump straight in without knowing anything about IT and just get into helpdesk and start "doing" (my cousin is doing this right now, did some bench work and got a call center job) but generally I think people get a cert or a few, study some things so that they go into IT knowing a bit of what they are doing.

    I learned this during those years.


  • Service Provider

    It was during that time that I did my MCSE+I and most of my CompTIA certs, for example. The entry level certs that helped to get me into the field. Coming from software engineering, back then, did diddly for getting you work in IT, even with a Solaris administration background!



  • @scottalanmiller said:

    @Dashrender said:

    And while I do believe all of that to be true from your numbers, It's still going to cost a company more money than self hosting.

    So you believe and don't believe at the same time? Why will it cost more? I know companies that have saved money doing it.

    So there are two parts here.

    The simple hosting of the server in the DC with power/HVAC/etc would probably be less. But only when comparing a sorta like (it's nearly impossible to get the same levels in house as it is in a DC) setup in house, which nearly no one does or tries to do.

    Of course I write all of this and recall a word you used - critical. You said

    Using a datacenter for hosting has been a standard since the 1990s and the idea that you never run critical apps in house since the early 2000s.

    So with that in mind, my associate left employ of that company about 7 years ago, so I guess he got out just about the time this would have been filtering down to his level.


Log in to reply
 

Looks like your connection to MangoLassi was lost, please wait while we try to reconnect.