How to pass-through connections on an intermediate Riverbed SteelHead

On Technology, ,

The Riverbed SteelHeads (SH) is a platform that, simply said, speeds up connections across the wide-area network (WAN).  Riverbed solutions improve network performance and thus user satisfaction and productivity, enable storage and server consolidation into the data center by making application performance across the WAN feel like the server or storage is still local in the remote office, and provide a visualization platform for applications needing to remain in the remote office while still being managed centrally, all while lowering ongoing operating expenses by avoiding bandwidth upgrades and requiring fewer managed devices.

One Riverbed customer had a question about how they could integrate two Riverbed optimized networks; one they control (A&B) and another, the do not (C&D).  They desired to keep the Riverbeds from each set of networks separate.  My response was posted to a technical forum but I also include it here since I found it very interesting.  It assumes technical proficiency with the Riverbed SteelHead platform, so unfortunately it will not be appropriate read for those not familiar with the workings of the product.

Using their diagram, let’s take a look at some of the technical requirement around how this would work:

SH_Peering_Rule_Example

In this scenario, networks A and B are optimizing via auto-discovery by Riverbed A and Riverbed D.  Networks C and D are being optimized by a fixed-target rule on Riverbed C pointing to the out-of-path Riverbed B.  The question was asked, if we don’t control networks C and D, what needs to be done to ensure that Riverbed A does not participate in optimization between network C and D.

If optimization is indeed occurring between clients on network D and servers on network C via a fixed target rule on Riverbed C pointing to Riverbed B, no peering rule is necessary on Riverbed A to avoid participation in optimization since peering rules are not invoked unless an auto-discovery probe is received.  Fixed-Target rules do not initiate such an auto-discovery probe since fixed-target rules hard code the Riverbed to peer with.  Furthermore, the optimized connection created by that fixed target rule from Riverbed C to Riverbed B is made on TCP port 7810 which is included in the default pass-through rules on Riverbed A for Riverbed protocols.

However, if we want to make sure that Riverbed C never peers with Riverbed A (due to sizing constraints for example), since both are in-path and may auto-discover each other, we will need both a peering rule and an in-path rule on Riverbed A.  Riverbed C may be configured not to perform auto-discovery, but unless we have visibility into its configuration, you can’t be sure.

To handle inbound connections from network D to network C, we would put a peering rule on Riverbed A, matching Riverbed C’s in-path IP address.  This ensures that a connection coming from a client on network D to a server on networks B or C would not be optimized.  In essence, the peering rule on Riverbed A says to not respond to auto-discovery probes from Riverbed C for any networks behind it.  We can also be more selective in the peering rule by using networks’ subnets instead of a peer IP.  For example, we could allow connections from clients on network D to be optimized (by Riverbed C and Riverbed A auto-peering) for network B but not network C.  We can think of a peering rule as answering the question, “what do I do when I receive an auto-discovery probe from another Riverbed?”   If we don’t want optimization from network D to network C, but we do want optimization from network A to network C, the peering rule on Riverbed A would be specific to a peer IP address of Riverbed C.  If we don’t want anyone auto-peering with Riverbed A when going to network C, the peering rule would use a destination subnet of network C.

Now, for outbound connections being initiated by clients on network C, due to the way server-side out-of-path works, outbound connections from network C will never be seen by Riverbed B. However, they will be seen by the in-path Riverbed A and thus will receive an auto-discovery probe. Other Riverbeds will respond to that probe and thus cause those connections to be optimized.  So, for a connection being initiated on network C going to network D, a probe would be generated by Riverbed A and that probe would then be seen by Riverbed C and they would optimize that connection.  Thus, if we don’t want Riverbed A and Riverbed C peering up for connections from network C to network D, we must put an in-path rule on Riverbed A to pass-through connections from network C to network D.  If we don’t want Riverbed A optimizing any outbound connections from network C, that in-path rule would just match the source address of network C.

In summary,  if we are just trying to keep Riverbed A from peering with Riverbed C, all we need is:

  1. An in-path rule on Riverbed A passing though connections to network D.
  2. A peering rule on Riverbed A passing probes from the peer-IP of Riverbed C.

This scenario still allows optimized connections from network A to network C or from network D to network A.

If we are instead trying to prohibit all optimization when networks A or B communicate with networks C or D, it is a little more complex.  To do that we need:

  1. An in-path rule on both Riverbed A and Riverbed D passing through connections to network D
  2. An in-path rule on Riverbed D passing through connections to network C.
  3. A peering rule on both Riverbed A and Riverbed D passing through probes from network D.
  4. A peering rules on Riverbed D passing through probes from network C.

 

 

Deeper Movies

George, Idea Polish Level: 1, My Ideas,

Have you ever watched a movie and thought that even if it was good, it was very shallow? For example, most real villains don’t do evil for the sake of evil. Something that I thought would be interesting would be if someone took a simple movie and made it a lot deeper. For example, they could show the villain’s back story, have the hero not be able to take down 20 ninjas by himself, have there not be 20 ninjas in the first place when they can use guns, etc. Just get rid of all those “that would never happen” things and replace it with something more interesting. Maybe retroactively make the movie into a book, since they say “the book it always better than the movie” (not including those little movie picture books for children and the like).

~ George

How To do tan^-1 on a calculator

George, miscellaneous, Miscellaneous Posts,
First off, I’m assuming that your calculator has three things: a “tan” button and a “tan-1” button. You do not have to know where the “tan-1” button is, but if it is a scientific calculator then it probably does have it. I’m also assuming that you know what the “tan” and the “tan-1” buttons do, so this will not explain that.
______________________________________________________________________
Your calculator should have a button labeled “tan-1”, although usually you have to hit the “2nd” key and then the “tan” key. If you are using the calculator that comes installed with Windows 7 (at least I’m pretty sure that’s where it comes from, if someone could verify that please) and already have it set to scientific calculator (hit the “view” button in the top left corner then select “scientific”), then you hit the “Inv” key (short for “Inverse”, I think) and the “tan” button will turn into “tan-1”.
Also, make sure that the calculator is set to degrees (or radians), because it is extremely frustrating to get the wrong answer and have no idea why. To set that there is often a button that says “DRG”, with which you can cycle through degrees, radians, and gradients, using arrow keys if your calculator has them. On the Windows 7 calculator there are three buttons in the top left, beneath where the output is shown. Click the one labeled with what you want.
Finally, if you already know how to get the “tan-1” button but don’t know how to type it in, most fancy calculators (the $10+ ones) require you to input the items in the order that you would write them (e.g. type “tan-1” then “500” then “/” then “1001” then “=”). However, almost every other calculator that I use (including the Windows 7 one and calculator apps) do it similarly, but certain functions (such as square root, log, and tan) are done after you finish typing the expression (e.g. type “500” then “/” then “1001” then “tan-1” then “=”).
Of course, all of this also applies to the “sin”/”sin-1” and “cos”/”cos-1” buttons, but I was specifically asked about the “tan-1” button, so that’s what I went with.
 ~ George

World Idea: No Passing This Point

George, Idea Polish Level: 2, My Ideas, ,

Imagine there was a civilization which has a mysterious floating point in its center. No one knows what it is, but if anyone goes outside a certain distance from it in any direction they quickly die due to unknown causes. However, if they make it back into the radius before they die they are completely returned to normal. This society is extremely advanced, and has managed to make contact with other such societies with their own mysterious floating points. Due to this communication, they learn that all the bubbles of civilization are different, some are larger, smaller, expanding, shrinking (uh-oh), in a desert, on an island, etc.

There are a couple of different ways that the “outside” could be portrayed.

  • It could be as if everyone was on Earth, but if any living thing left the spheres of protection then they’d die. In that case, the inhabitants would be able to see things such mountains and oceans outside their sphere but not be able to interact with it.
    • If this was the case, it seems likely that the spheres were made for the protection of life, because whatever was outside of them was extremely dangerous. Too bad they don’t know how, when, or why the spheres were made
  • Another possibility is that only humans are forced to stay inside. If anyone left then they’d die, but the pets, livestock, and plants could survive outside the invisible sphere.
    • A world such as this, with humans being the specific target, is more suspicious, and seems to be more likely to be caused by sentient life who were trying to keep humans (or specific types of humans) captive. Aliens maybe?
    • Also, if the spheres become crowded, this could lead to some interesting farming techniques, which might involve leavingall of the livestock outside of the radius until the farmers needed them. This way they don’t have to worry about leaving space for food as the population grows
      • In fact, if the place is highly advanced they could do almost all of their work outside of the radius using drones. They could have everything we have today by building factories far away from the sphere and having farms (or maybe no farms, if no life can leave the sphere) cared for by robots. The exception to having everything we have would be space to run around in, which would probably lead to a booming industry for virtual reality, maybe even inter-sphere competitions if they could get it to work
  • Another possibility is that it’s a little planet/moon/asteroid of sorts (something big enough to have about one city, but not much more), but no one can get far from the surface. Maybe the dot would be at the center of the planet. They would still be able to see the stars and do astronomy, but they wouldn’t be able to leave
    • Of course, while that wouldn’t be too hard on our civilization, due to lack of resources for interplanetary missions, this is an advanced civilization which probably could get between planets with ease. Not being able to leave could be a real problem for them
  • A fourth possibility is that outside the sphere is simply nothing, and if anything made of matter leaves it simply disappears. It’d be a sort of “edge of universe”, and there would be absolutely no way of expanding beyond the sphere. Perhaps things that aren’t matter* could still leave, such as light, but it’d also make sense if nothing that we regularly interact with could leave; after all, it’d be the literal edge of the universe, there’d be nowhere to leave too. I don’t know how the different “universes” would communicate with each other, but it’d be extremely interesting

~ George

* matter: things that have mass and take up space

Change log of October 26, 2014

Change Logs, George, Miscellaneous Posts

At the end of every day (or, at least every day that I change something), I make a post that highlights all of the changes and/or posts that have been made since the last change log. This way I can update previous posts and actually have the changes be found, plus readers will be able get easy access to all the new posts without scrolling down too much. Note: Since this post will be the last thing I post each day, it is less likely to be properly cleaned up, as I will be rather tired at that point

New posts:

Major Changes:

  • none

Minor Changes:

  • none

Of Special Note:

  • This change log is a day late because I was really tired last night and didn’t feel like making a change log

Certified Brilliant:

https://www.youtube.com/watch?v=kTcRRaXV-fg

  • I believe that if you haven’t seen Abbott and Costello Who’s on First then you are truly missing something in life. They say that actions speak louder than words, so rather than just recommending the video I’ll actively make it easier for you to see it 🙂

 

~ George

Culture Idea: Chance Based Currency (Part 2)

George, Idea Polish Level: 4, My Ideas, , , ,

The next part of my explanation of the Chance Based Currency idea!

(See part 1 here: http://evansforever.com/culture-idea-chance-based-currency-part-1/)

The first problem I’d like to address is that it would be very hard to deal with large sums of money.

  • Using high felix dice, like a 100 felix die, would make the seller risk gaining only 1 F in exchange for an item. On the other hand, the purchaser is also risking paying 100 F for the item. On average, the seller would make about 50 F per sell, but that’d only be after a while.
    • As said in the first post, all the seller can do to decrease the risk of having a low F roll is to refuse to accept high felix dice. However, if the government is like ours (the US), then it’d be against the law to not accept currency the government has decided has value.
      • In the US, if someone tries to pay you with legitimate US currency (and you’re in the US), unless you accept the currency the buyer has the right to take whatever it was that he was purchasing without paying you. The dice government might have similar laws, or they might only have laws for specific dice types, or maybe none at all
  • Another problem would be having to make a 100 sided die. Perhaps the government would make dice that go up in intervals greater than 1. If so, then a 100 felix die could be ten sided, going up by ten, or 20 sided going up by five, or maybe even 5 sided, going up by 20. This would clearly affect the formula to calculate average F, which would become as follows: Divide the max felix of the die by two, then add (the lowest value it can roll*1/2).
  • This would put the 10 sided 100 felix die’s average F at 55, the 20 sided 100 felix die’s average F at 52.5, and the 5 sided 100 felix die’s value at 60! Wow, and I thought this system was interesting already, but with accounting for minimum values as  well  as  maximum values makes this all the more fascinating
    • And what if the die didn’t go up by the same interval each time? What if you had a 100 felix die, but it had only 6 sides, which were 1,2,3,4,5, and 100? It’s still 100 felix, because felix is the potential roll, but it would have nowhere near the purchasing power as the other examples given.
    • If this were the case, the way you’d calculate the average F is by adding together all the sides and dividing that number by the number of sides on the dice. For example, 1+2+3+4+5+100 =115, 115/6=19.1666667, which is the average F of that die.
    • This method could work with other dice, for example 1+2+3+4+5=15, 15/5=3, which we already know is the average F for a 5 sided 5 felix die (man, now I have to say how many sides it has also…), but it’s harder and more time-consuming than the normal calculation.
    • Another thing I thought of about this idea is that it’d provide another way to gamble without any felix loss; roll an “x” sided die which has all of its faces equal to zero but one, which would be some high number. Perhaps the government would earn extra money by selling dice such as these to gambling houses.

That’s all for now

~ George

Change Log of October 19, 2014

Change Logs, George, Miscellaneous Posts

At the end of every day (or, at least every day that I change something), I make a post that highlights all of the changes and/or posts that have been made since the last change log. This way I can update previous posts and actually have the changes be found, plus readers will be able get easy access to all the new posts without scrolling down too much. Note: Since this post will be the last thing I post each day, it is less likely to be properly cleaned up, as I will be rather tired at that point

New posts:

  • Interesting Observation: Average Day considers what we say we like to do and what we actually like to do
  • Finally! Another Idea Polish Level 4! Culture Idea: Chance Based Currency is something I actually started about a month ago, but I kept thinking I couldn’t publish it until it was finished. However, I decided that it’d make it easier on both the reader and the writer to have it split into smaller sections that don’t take as long to read (or write 🙂 ), so that’s what I did.

Major Changes:

Minor Changes:

  • I reformatted the Cool Websites page, Extra Cool Websites page, Special Websites page, and Ponderings from Math Class post so that each item in them were individually numbered.

Of Special Note:

  • Subscribe to the blog to get email alerts whenever a new post is posted!

Certified Brilliant:

 

~ George

Culture Idea: Chance Based Currency (Part 1)

George, Idea Polish Level: 4, My Ideas, , , ,

What if there was an economy which had legal tinder that was made of dice? I’m going to call the currency Felix (“lucky” in Latin). For example, a two felix “bill” would be a two-sided coin, a six felix “bill” would be a normal 6-sided dice, and I don’t know how they would work with hundred felix “bill”‘s. When you pay for something, you pull out your dice and roll them. Whatever value they land on is how much they are worth for that transaction; you could have a 100 sided die, but if it lands on 1 then it’s only worth one dollar. Perhaps when paying something, you have to bring out enough dice to have the price be halfway between the minimum you could roll and the maximum you could roll (I’ll explain that more later), and after the cashier checks to make sure that everything adds up properly you roll the dice. You pay whatever value that comes up is, whether cheaper than or more expensive than the original price.

This could lead to an interesting treatment of the value of money. Here’s some math to explain: You’re buying a new hammer that is worth 5 F [F is absolute money (after the die has been rolled), and felix is potential money (pre-roll), e.g. a 10 felix die rolls 5 F]. If you have a 5 felix die, you still probably wouldn’t be able to afford it. This is because, on average, the die wouldn’t roll a five, and thus, on average, the seller would lose money. The seller doesn’t want to lose money, so he wouldn’t sell the hammer for a 5 felix die.

The way you’d calculate the average value of a die is to divide the top value it could roll in half and add 0.5. The additional 0.5 is because the die can’t ever roll zero, so it’s not the halfway point between the top value and zero that you’re looking for, it’s the halfway point between the top value and one. This would place the average value of the 5 felix die at 3 F. To get you’re hammer you need an average of 2 more F, so if subtract 0.5 from 2 and double the outcome you see that you’d need a 3 felix die to complete the transaction (I assume that a culture based on this currency would figure out how to make a three-sided die).

So now you have two dice which have an average F of 5, enough to satisfy the shopkeeper. You roll your dice. The 3 felix die lands on one, good for you, but the 5 felix die lands on 5, for a total of 6 F. Your heart sinks. The shopkeeper happily pockets the dice, having earned an extra F, and gives you the hammer.

Later, your friend sees the nice quality of your hammer and gives you a 9 felix die to buy him one. The interesting thing about this situation is that a 9 felix die also has an average value of 5 F, even though a 5 felix die + a 3 felix die = 8 felix. The difference is that every die has a minimum roll of 1, so the minimum F for two dice is 2, meanwhile the minimum F for one die is 1. The added price of 1 felix accounts for the added risk to the shopkeeper of 1 felix.

You visit the shopkeeper again, and he seems a bit worried about your 9 felix die, but doesn’t stop you from using it to pay. You roll, and his fears are confirmed. The die landed on 2, giving it a value of 2 F, 3 F less than the asking price! You can see that the shopkeeper is upset at being shortchanged as he pockets the die, but you’re elated. You can’t wait to get the hammer to your friend so that you can tell him what a steal you got it for.

In this system, most sellers would always want to be paid in the highest number of the lowest denomination dice they could get, at least for the more important deals,  so that they are guaranteed at least a certain amount of F, even though the fewer dice that are used the higher the felix value is. Some shopkeepers wouldn’t allow dice with too high of a felix value to be used to purchase items, meanwhile others might make a sale by requiring the average F to be less than halfway between the top value and one. Gambling would be easy in this culture; simply have both players roll a 100 sided die (or whatever they use instead) and switch dice. One might roll 50, and the other might roll 20, who gets 30 F while maintaining the same amount of felix.

After all that though, the only more flawed currency system that I’ve seen anywhere (not that I’ve looked for any) was this one:

Alternate Currency

(Source: http://www.explainxkcd.com/wiki/index.php/512:_Alternate_Currency)

I can’t decide which flaws I should go over first, but this post is too long already, so I’m going to split it into a number of smaller parts focusing on specific problems with this system and addressing them. These I will write and publish those parts until I’ve gone over this idea thoroughly, or I’m tired of it.

~ George

Interesting Observation: Average Day

George, Idea Polish Level: 1, My Ideas,
When someone asks you what you like to do, you usually don’t reply with the things you do most often, whether because they’re not fun (school, work), everyday (eat, sleep), or boring to talk about (watch TV, use the computer, play outside, etc., although they can be interesting to talk about, they aren’t very noteworthy). The more usual reply would be something more specific and interesting, like I would say “Swim Team and Piano”, because those are more interesting replies than “using the computer”. In general, it’s the highlights that we talk about, because for the average person the average day is merely average : )
 .
(Yes, I wrote this whole thing primarily so that I could say that last line 🙂 )
 ~ George