Saif shaikh

Hi. So this is, uh, Sheikh, and I, um, I'm I'm a b graduate from electronic telecommunication. And, uh, you know, more of a brief about me is I right now, I am, uh, very fascinated to, uh, network automation, network yeah. I'm I'm I'm a network Cisco network guy. I'm a proper network guy by my heart, but, uh, I started my journey with network engineering because this is something that really drive drives me crazy. Uh, and you talk about network to me for for for for whole day long. I'll I'll I'll, uh, I'll I'll do that for you. But, again, uh, I started my journey with, uh, with a company called, uh, MicroSense that had a lot of wireless projects. And, uh, the post that, I got an opportunity to work in a BFSI domain where I dealt with, uh, lot of stock markets like NSE, BSE, MCX. And, uh, I was part of designing some casting for a vendor. So I was not working directly with NEC, but I used to take the feed from them. And then there were some, uh, vendors that we used to work with, like, Grinksoft, Greenland Investments. And, uh, so so the idea was these guys were making in software where you could sit and, uh, trade on different exchange so you don't have to go on different exchanges for trading. So this was a consolidated platform where you could trade. Uh, so so and, anyways, the idea was not for me not to trade, but I was a part of an infra team who would deal with not only networks, but, uh, but also understanding how, uh, rack and stack and how how things go from ground to hub. So, you know, dealing with Linux servers, ensuring that, you know, they are PTP enabled, NTP enabled because those those were real time real time applications, and I really enjoyed working there. And cutting it short, I now, um, I also been a a tech engineer in Cisco. Um, and now I'm a Cisco employee who's working, uh, for a product called Skylight Platform where we are working with all the service provider in India and outside India who who is monitoring the network on on a minute accuracy. Uh, so so we're leveraging the protocol called TRAM. So we have set of elements, uh, in Cisco in in Cisco Skylights product, um, where you're working for a full stack observability. And the idea is to have ensured that, you know, you're delivering, uh, you Uh, you know, we're not helping the the service provider to make their network fast, but we're helping them how can they make fast. Yeah. That is all from my side.

Yeah. So, uh, describing process of of managing and implementing SD WAN technologies. Uh, okay. So, yeah, uh, I think, uh, SD WAN is is is the key right now for an intent based networking these days. Uh, you are, uh, no, with an intention, you are, you know, you can have a proper flow. I mean, anyways, SDN ever since SDN has has ever, uh, hit the market, we, uh, we we we are segregating the control plane with the with the data plane so that, you know, ensuring that a single we have a centralized control over the network. So, anyways, the the question I'm just repeating it once again that, you know, how would I manage an SD WAN technology? So, you know, my my head office would have all the set of components that, uh, that is that is the that the SD WAN provides, the Wiptila provides, uh, my my VBOND, my VMANAGE, my, uh, my my VHS on the different different of my location. So so I definitely would need and see a cert CA server that would issue my, you know, server's certificates, and then ensuring that I have a PKI done. And once that is done, that is the only hectic thing that we do. And then once that is done, uh, we just need to onboard all my v edges. And I have 2 different transports, one MPLS, 1 Internet, and maybe 1 point to point connectivity, 1 MPLS. Yeah. MPLS, I already said. So we're ensuring that, okay, auto we we know that how much the application is, uh, is the severity of the application and how much it is, uh, important for an application not to have any any any any bad KPIs to ensure that we have an hidden good KPIs, and then we are using the the the one thing that is that that is the best. Anyways, uh, if you ask about how would I, uh, you know, process and man manage it, uh, it is pretty simple once it is installed and once it is, uh, you know, once it is up and running.

So how would you optimize in QS setting for a few IP infrastructure, uh, to mitigate a packet loss or repeat usage? So, uh, why is we a very, um, to be very honest, I've really not worked on, uh, uh, any VoIP related applications. But but, nevertheless, I I have I have a fair idea about it. Um, so so, basically, your VoIP, uh, traffic, your voice voice traffic is pretty, uh, you know, really faster than your data. Your data can still wait, but your voice traffic cannot wait. So ensure so so so ensuring that I give a a high priority, high DSCP values to, uh, you know, to put these packets, you know, marking to my QS is whenever I implement a QS, I would have 3 things in my mind. Right? I mean, first, I have to mark classify it. I have to mark it. And then I have to, you know yeah. I have to classify and mark it first, very first thing. So when I, uh, ensure that, you know, I'm classifying it on the source side. Okay? That, okay, this is my voice traffic or this is not my voice traffic. Uh, if it's not my voice traffic, I'm not marking anything. Then I decide on, okay, how do I mark it? So the highest, uh, if I give that, okay, do not ever drop these packets because these packets shouldn't get dropped because, you know, if this gets dropped, this is useless. Even if you retry it, it won't make sense. So whenever these packets are in the queues, process it as soon as possible. So so so the with the proper marking, uh, then I can, you know, implement those templates in all the notes from from my point a to point point b all the way, and then, uh, that's all. I mean, I think if I have enough bandwidth, I would never need a QS. This this is my understanding, because I would never drop a packet. But if if I have a, uh, if I have a bandwidth crunch, um, I would rather insist, uh, the the the enterprise to, you know, to have enough bandwidth so that, you know, we we never drop anything.

So how would I seamless run do a transaction to transition from okay. From from Cisco to Brocade there? I mean, I would rather see that the the the placement of that Cisco switch, uh, which is, you know, what it is really doing there. And, uh, is it acting is it acting as, uh, as a basic bridge of what it is doing? So I would I would first list down all the, like, the, uh, you know, all the functionality that which is providing. And then, thereon, I would, you know, go open up my Brocade uh, command line or, you know, take a help from, uh, Internet or so that okay. How would I configure you know, suppose if I'm doing an STP in that switch, uh, message switch or if I'm doing a layout in that switch or QoS or so, whatsoever. So I would I would just list down all those things. And then I would go back to that bucket switch. And then I would rather, uh, study about it that, okay, uh, what what things how do I configure STP in network routing in this

So I have never used the poll for scripting. I'm using Python. I'm using bash. So, um, it's I'm I'm not limited to bash and Python. But, yeah, uh

Uh, discuss the implication of COBT, COBIT, uh, framework. That. Okay. I'm I'm really sorry. I I don't know what's your, uh,

Can you explain the different voice? Yeah. Yeah. Yeah. So definitely, I'd love to speak on this for long. Uh, so your OSPF is, again, a, like, say protocol, who something who knows? I'm sorry. Hello? Yeah. So OSPF is something that, uh, would, uh, you know, the link state protocol who knows the whole topology, the whole the graph of the pro of the of the network, and then it would, uh, know know what is happening right there in that area, uh, not across the area. I mean, we would definitely segregate it, uh, once I you know, once we make different different areas. But at least in the area, I know, uh, you know, what is happening, uh, which link is connected to what link. So the idea is to have when when you say that I it's it's a link state flavor protocol. That means it is, uh, it is something that it could keep update about the whole network, whole topology, assuming that we are in the same area. Uh, even if you're not, I at least know that go to that area or go to that that particular node, uh, I have to reach out to my ABR or something like that. Uh, the best thing about OSPF is it is scalable, very scalable. You don't need to think about, uh, you know, when I compare it with the IGRP, my routing table is not I haven't at least, uh, what do I call it? I at least have an option to shorten down routing table or something like that. OSPF is is, uh, is something that is, uh, when you compare it with other routing protocol with mostly with you, compare it with EHR, or BGP, uh, because you will not compare it with ISS because it's it's 1 and the same, Uh, 1 and the same in the sense, uh, the it's both linked state protocol. So I'm sure comparing link state with distant spectrum. So so so the only difference, uh, if you talk about OSPF and AIGRP is that, you know, OSPF is pretty scalable for for, uh, uh, large enterprises, large service providers, and uh, it can it is it is good for traffic engineering. It is very scalable. And apart from, uh, it's not, uh, that doesn't mean that is bad. It completely depends on network to network. And, um, yeah. And that is it.

Yeah. So your VRRP being, uh, you know, the the f one of the FHRP, you know, you're the 1st stop to go out of your network. Uh, you know, ensure if your gateway is dead, uh, you will not route yourself. Right? I mean so so the idea is to ensure that your your your gateway, your Tor, your top of the rack, uh, device is always up and running. Uh, so so so we are it really helps you to ensure that if we have a virtual IP, we we, you know, we are, um, literally, uh, ruling out, uh, the link failure as well as the physical, uh, box failure. So so we are actually being one of the protocol that would help you to ensure that your top of the rack is always up and running, and that would ensure that your network is always up and running all the time.

Comprehensive project Depends on how many how many routers am I am I am I replacing. And, uh, so it is you know, I would rather take, uh, first of all, I would plan it up at how how many numbers of sites I have and how many routers I have. I would create, create all those configuration in one go. Okay? And then, um, I know I would I would take a downtime and then push it. So the timeline, uh, I I'm not sure really what the this question is asking me. But, uh, I mean, if I can sense it that if if if it's about time line, if you're asking me what is the time line that I could see, I think I really need to know how many how many number of devices there in the, uh, in in in the network. And uh, if if I take a common example of if I have 3 or 4 sites, I think it'll at least take 2 hours to for for 1 site to do because, uh, I would I would perform a test, uh, pre check and post check. Um, but but in doing all those things in the production live, I would rather do it in the test lab first than how much time it is taking for me to to remove the VRRP and ensuring that, you know, my existing service is not impacted. But, uh, I would rather have to take a downtime because I, uh, if I'm if I'm removing VR reconfiguration, my all the devices beneath is going to, uh, you know, talk to that virtual IP, which I have, you know, given you. So, yeah, the idea is to to have, uh, the number of sites that we have. So if it's about 1 site, and at least take we will at least need 2 to 3 hours to ensure, uh, with pre check, post check,

So, uh, the key, key element to focus on network modifications are, you know, definitely test bits. Uh, every changes that I've made, we have already, you know we we make sure that, you know, we're we're checking in the in the labs first. We're doing the test lab, uh, test a bit first and then making the modifications, um, listing down all services that is gonna get impacted. That's first of all, listing down all the services that that site or that network is, uh, providing and then, uh, then then ensuring that, okay, those are not getting impacted, uh, and what what other redundancy that I can create. Uh, Yeah. That's that's all.

Uh, SD WAN technology, as as I said, uh, impacts on an architecture. Um, I think, uh, it's it's an intent based networking. You know? You you know that, you know, way of pocket package gonna flow. Uh, your your control plane has become so sharp that it knows, um, that, okay, if mine this particular, uh, traditional run link, which is taking, um, approximately, uh, no the KPIs are not good or something like that. I my application needs, uh, anything below 1 m 2 MS. So I would rather route it through that way. And, uh, and that benefit that it provides over MPLS, you know, maybe MPLS. Uh, I think I don't I don't think that, you know, uh, you it it completely depends on the provider. Again, if your MPLS is faster than your your WAN, uh, traditional wired architecture, then then then it's good.