From GhostCMS to Hashnode - slightly controversial Computer Science |
- From GhostCMS to Hashnode - slightly controversial
- how much is the workload if i major in statistics and minor in cs?
- Artificial Intelligence Helps to Resolve Long-Running Astrophysics Debate on Supermassive Black Holes
- Performance of local filesystem like ext4 compared to a clustered file system like ocfs2
- Confused about the utility of router(host) loopback address other than for testing purposes
| From GhostCMS to Hashnode - slightly controversial Posted: 06 Aug 2021 02:58 AM PDT |
| how much is the workload if i major in statistics and minor in cs? Posted: 05 Aug 2021 08:15 PM PDT can yall give me the cold truth? how much free time will i have if i major in stats and minor in cs? [link] [comments]  |
| Posted: 05 Aug 2021 08:16 AM PDT |
| Performance of local filesystem like ext4 compared to a clustered file system like ocfs2 Posted: 26 Jul 2021 10:16 AM PDT I know this is not a fair comparison, but I would like to know how much worse ocfs2 performs when compared to ext4. Would really appreciate if someone posted relevant research papers. [link] [comments]  |
| Confused about the utility of router(host) loopback address other than for testing purposes Posted: 24 Jul 2021 05:09 AM PDT Hello everyone, I'm learning networking concepts and I'm at loopback IP addresses. As the title suggest I'm having some difficulty understanding the loopback interfaces particularly why they are used as source and destination interfaces when we send data. I will try to articulate my question as best as I can please excuse me if I'm not able to do so as I'm very inexperienced in this topic :D. I will use an example to better explain it. I highlighted the questions in bold towards the end of the post. I understand that loopback interfaces are used for testing purposes. For example if I would want to check that the tcp/ip stack running on my OS is working fine I could try to send a message to myself and see if I receive it in the application layer. I can understand that. But I've seen that loopback addresses are also used as destination/source addresses for example when a device wants to send another remote device some data it can specify it's own loopback address as source ip address and the loopback ip of the receiver as the destination IP. Even more, if I understood correctly a device can send some data to another device which is directly connected to it but because the loopback ip has a /32 mask the device will be seen as a remote device and the routing will be done on layer 3(network layer) instead of layer 2(ethernet or some other protocol). From what I've read not only is this possible it is actually best practice to do so.As I understand this could help if we had a topology such as this one below with two routers connected to each other: R1 and R2 are connected to each other via 2 interfaces.If router R1 tries to send some data from 10.1.2.1 to 10.1.2.2 interface of router R2 but one of these interfaces fail for some reason then data would be lost. However if we try to send data from loopback interface of R1(say 1.1.1.1/32) to loopback interface of R2(2.2.2.2) and the physical interface specified above fails the router R1 will know to use the other hardware interface for the data transfer(provided the needed information is specified in R1's routing table of course).As I understood this is the main utility of using loopback interface as source/destination. But is this really necessary? Can't we just specify both hardware routes from R1 to R2(10.1.2.1-10.1.2.2 and 10.1.1.1->10.1.1.2) in R1's routing lookup table and if the first link fails the router will know to use the second one?Or is there another advantage of using the loopback address that I'm not seeing? Also when a device wants to send some data to a neighbour device via their hardware interfaces the routing happens purely in layer 2(e.g. ethernet) without the ip address being needed. But if we instead use loopback interface then there will also be layer 3 routing involved between neighbouring devices. Is this an actual advantage? Doesn't this actually take more time since the receiving device also needs to process the IP packet as well? I'm guessing the time it takes for it to do so is not exactly substantial but why would we ever want to use L3 routing where L2 works just fine? Isn't this complicating things without reason? Please help me understand this. [link] [comments]  |
| You are subscribed to email updates from Computer Science: Theory and Application. To stop receiving these emails, you may unsubscribe now. | Email delivery powered by Google |
| Google, 1600 Amphitheatre Parkway, Mountain View, CA 94043, United States | |
No comments:
Post a Comment