Demystifying the Role of The Network Designer with Mohamed Radwan – ZDNP 074

ZNDP 074: Demystifying The Role of The Network Designer with Mohamed Radwan

Zig Zsiga:
[0:00] Demystifying the role of the network designer with Mohamed Radwan episode 74.

In this episode, I want to help you find the role that you are meant to do in networking I am highlighting all of the different roles we have like network engineer Network designer Network. Even the new DevOps engineer roles and the automation specialist roles. So for me personally I would rather work in a position that I loved, then make four times the salary in a position I hated and that took me away from my family I always say this find the position you love and embrace it.

Today to help me on this goal of helping you find the role you are meant for in life in your career in networking, my good friend Mohamed Radwan is joining me. Mohamed is a leading Network designer and architect he's a fellow CCDE and CCIE. Mohamed just recently was promoted to a CTO, that's Chief Technology Officer which is outstanding. Mohamed joins me today to highlight and demystify the role of the network designer so stick around here we go.

Welcome back my Friends, Nerds, Geeks, and Ziglets out there we have another episode of the Zigbits Network Design Podcast where Zigabytes are faster than gigabytes. As always, we strive to provide real-world context around technology.

I hope everyone is doing great Zig Zsiga here and welcome to episode 74 of the Zigbits Network Design Podcast. Once again my name is Zig Zsiga, I'm here to help you with network engineering, Network design and network architecture, and one topic that we have to cover today is the role of the network designer. And to help me today with that is my friend Mohamed thanks for joining and how are you doing today buddy.

Mohamed Radwan: [1:54] Thanks Zig, it's always a pleasure to meet you and talk to you it's today I'm really happy because I was telling you before the life that, today was a special day for myself as my daughter she's just move up in her career she's becoming like 82 referee in basketball and today was her first match by herself so I was very proud of her and I'm very happy and exciting to be with you today.

Zig Zsiga: [2:19] Hey man that's awesome right we gotta we gotta get those wins out there and especially in this time you know I mean we're in the New Year today this is going to be this is airing in January this is the new year and we're trying to forget about 2020 but we really got to think of those positive things and that's great that's huge all those family things that are going on so what sport it is again.

Mohamed Radwan: [2:40] She's a basketball referee.

Zig Zsiga: [2:42] Wow see that's cool and I played so side note just so everyone knows I played basketball on and off for a couple years and Junior High and high school and then I actually got cut because I suck at it my senior year in basketball so just some you know tidbit of information for all of you I suck at basketball I am a soccer player a traditional soccer player by heart.

Mohamed Radwan: [3:08] Great me too by the way.

Zig Zsiga: [3:12] Real quick totally off topic is always right so we talked about soccer where you are because you live in Sydney just so we're clear right you live in Sydney Australia, what do they call it football or soccer.

Mohamed Radwan: [3:27] Is equal to soccer as well.

Zig Zsiga: [3:29] See I didn't know that. And that's why I was asking I had no idea so that's cool Okay cool so hey again thank you for joining today I appreciate it buddy you know we've been doing a lot of things together over the last six seven months and I don't know why we haven't done a show like this before I mean I don't know why.

Mohamed Radwan: [3:45] Exactly what all of that period and just now that I really enjoy all the things we did together actually we giving back to the community is something you feel like a blast after doing that for the people and supporting our Beloved Community.

Zig Zsiga: [3:59] Oh yes man this is our role right we want to give back to this community. Every step of the way so and again so we're going to copy we're going to copy we're gonna cover the network designer roll right and honestly you're the first person to help me with this in this context demystifying the role of the network designer so as always I have a whole bunch of questions that we're just going to kind of run through is always and really the first one here is what is the role of the network designer?

Mohamed Radwan: [4:25] Yeah it's a really good question and a good start Zig so, I think the role of the toxin is the key when it comes to the process some people will say that mainly the network designer what he need to do is to plan and prepare, and do the due diligence required to complete a network implementation so I believe even if you are an old engineer like myself or even older the people 25 years old beautiful inspires years experience or back to use just run to the network they have an idea I want to configure ospf they just something that we can go and configure it and then they discover this is not the optimal way to do a project is not the optimal way to manage a network so.

Then the role of network design become critical and critical it's also a part of system design so if you design like any it it's very similar to designing your infrastructure your Microsoft clustered you database and so on and I believe it's required a lot of preparation you need to be very solid understanding your tools and after you do that. Planning do your due diligence and be like you like this word yourself, you and myself Zig coaches being like a bridge between the business business requirement and Technical solution to actually this is a key.

Zig Zsiga: [5:44] So you just took the words out of my mouth and you know we've been presenting on webinars for a while but that's what I always tell everyone to is we really are Bridging the gap between the technical side of things and the business side of things we have the speak both languages in my opinion. And it's different languages right you're the technical people care about the technical things and then their business people care about the business things and the wording is all different you know we're talking about Opex and capex the business side of things.

And in total cost of ownership and all those you know.

Business terminology really and then on a technical side were talking about ones and zeros and IP addresses and subnets and I don't know what OSPF and bgp and you know exactly so okay cool cool now on that note.

What do you do as a network designer today?

Mohamed Radwan: [6:38] For sure even a now I moved up a bit so as we can see it's like a film it's like a cycle so it's made the made the the role or the function May differ but in many companies you have that cooperation then implementation then design then architect and maybe then you go, the management and so on so as a designer usually it may differ from industry to another industry or from company to another company but your main role designer or / architect is to engage early with the customer understand the requirements.

And then start building multiple documents so for my from my long experience I believe some of the clear documentation is critical so even when you start meeting the customer one of your main role is to do after meeting maybe meeting the customer Gathering the requirement. I understand his need usually you in this phase which mostly the planning phase with the document either you.

The requirement document customer requirement documents or data Gathering report regardless of usually I believe usually in each phase of the project so after planning you prepare this document.

[7:47] Maybe then you move to the second phase which you starting the design so when you start the design you start thinking about something like how I can translate this business requirement into a high-level solution what what are my building block what is my architecture how many modules should I include in my solution and always you need to think about this design pillars like.

Is my solution secure is it scalable is that manageable and so on so I mean it's a look like passion because I love this network design I've been there as a designer for more than 15 years so it's required a lot of interaction there is a lot of documentation so you need to be solid when it comes to the communication you need to be solid when it comes to communication with the customer so this is just an example so,

Zig do you have like a comments or the answer your question?

Zig Zsiga: [8:41] Yeah it did answer my question but I think I think it'd be beneficial here for us to actually break this apart, you mentioned phases almost so I think. And I’m going to put some words out there as always right.

We have Phase 1 unless I don't know if it makes sense I would call that like Discovery or requirements Gathering or something along those lines that's a phase one does that resonate?

Mohamed Radwan: [9:05] Yeah exactly.

Zig Zsiga: [9:07] And so if I can take what you said they're right you're that phase one process as a designer you need to get into that that customer space or if it's your own environment and you have to do a full discovery of the business State what they care about but it also the technology State what's being done today and what are the workflows and how are things being done and what are the pain points all of that you're really just getting a good discovery.

Mohamed Radwan: [9:31] Hundred percent and I key Point here also for the discovery phase is identifying your stakeholders. As you mentioned where you have an internal project let's say you have your working on a company and I had that before and you don't like acid refresh or maybe adding a new data center so in this case you have internet customers we need to identify who is a stakeholder.

[9:52] Is your city oh maybe your project owner your CIO your security architect and then you go and meet with them and you're prepared like questionnaire then you start,

Gathering the requirement so this will be very important Fields first you need to gather the, I think I like the top down approach to you start thinking about business people process, another that used to only start touching the network and then even when you start talking about technology you start from Level layer 7 and also ideas we start talking about the application what are the application on the network how they communicate where the users are located and and then you go deep and deep to the lair to reach Network.

[10:37] Physical and after that maybe you can stop asking about the data center of the location the power the cooling so any over the year I believe most of us maybe we developed like a couple of questions or couple of questionnaires templates, which we keep modifying and using with different Technologies today I'm meeting like a customer he is looking for maybe he's a service provider so I got for my my brother the service provider questionnaire tomorrow I'm eating like a cloud customer looking for a cloud migration so I grabbed another question is so this this phase is very critical.

Some people they take it easy or over think it or so sorry just maybe don't give it the right amount of time and then this really affect the.

[11:23] The consecutive phases like the planning phase, the design phase and implementation phase.

Zig Zsiga: [11:28] Well that's great that's good okay so Discovery phase we do that we get the requirements we understand what's going on and then we move into the beginning kind of we have to process those things but you mentioned something in that Discovery phase that I think was important identifying those stakeholders.

And that is hyper critical and I've been saying that a lot lately it's my new buzzword I guess hypercritical but that that is super important because you have to know who you're doing all this for.

You have to know who is the line of the business owner the line of business owner that you're talking to you have to know who your stakeholders are so you can have those conversations and they're going to be the ones that can,

I tell you what the priorities are can if there's something that's competing priorities they can deconflict them and then they could also identify Financial items to so if they have funding, budget or they can say hey no this is an actual need requirement we need more money than because we have to do that.

So they have to be kind of in this process right.

Mohamed Radwan: [12:33] I agree a hundred percent and as you know Zig's some we're not worked working with machines so stakeholders is our people also humans so sometimes also this data Gathering approach some people do prefer to do it over multiple sessions so if I meet in person today maybe I'll give him a bunch of question they also give him some time to go and understand my question they need behind my question I tried to explain it to him and after that this on his answer maybe that's the second batch of questions so this it attractive approach for design I believe it's, it starts early in the project.

Another point I want to highlight here, which is very important and I noticed that and some companies are in some situations they don't really engage or involves the design team early in the project.

[13:26] Big mistake so sometimes maybe if you have a big company and they have a huge sales team maybe the sales team will go meet the customer understand the requirement and even decide on the hardware and after that the Engage The Design Team or the architecture team this is a big mistake and actually caused a lot of disaster which I and opposite for during my experience some of the project which we the Arctic team was engaged very early in the project even sometimes during the

RFP during the negotiation or even the first stages when it was it was just an idea it's a huge different to the project success and progress.

Through the life cycle of it.

Zig Zsiga: [14:12] No that's a good point because I find it I find it better now these days where design is brought in a lot earlier at the beginning phases but like if you go back 10 years ago it was always like a like a oh we were supposed to bring you in and the projects already done like it's.

Everything's implemented and it's nothing's working and that's why we're being discussed why that's why we're having this conversation is just a problem and so it's like all right mr. customer you didn't bring me in like I can't solve your problem now because it's live you put it in production and you designed I don't know you designed this for whatever requirements that you thought you had and there's actually you at least other five requirements that no one even realized and then, and then you're like well why didn't you tell us this you know what well you didn't bring me.

Mohamed Radwan: [15:00] There and it'll have to be the best surprise when you go to the customer and he discovered he already bought the hardware. And hatin yeah do your magic so I got this it routers and I need to build my network so yeah I don't have any switches but you have to have to manage and build me thy campus Network without switch is also on the so.

Zig Zsiga: [15:24] What's that saying like putting a square peg in a round hole or whatever like that. You're trying to make it happen because at the end of it and you're putting on your you're putting yourself in a spot and they're putting you on the spot and it's like well it's not going to happen you didn't get switches like come on.

Mohamed Radwan: [15:41] This is unprepared sfps or Sonia.

Zig Zsiga: [15:45] The SFP is not a hard fix like that I mean You Gotta Buy one right but like I mean those are literally those are little there's like cables in most cases there might be a bit more expensive than cables right but but then it's like if you didn't get like Hardware like that's a problem right that's a that's a you know or he designed a solution that's not optimal.

Um that's a problem too so yeah that the lesson there and I'm we're being I'm being long-winded it's me make sure you're including design at the beginning of the process for sure.

Mohamed Radwan: [16:14] Hundred percent hundred percent is a golden tip.

Zig Zsiga: [16:18] All right so so then let's move on to the next phase right so we here we have our requirements we know what they're looking to achieve we know what they're doing today we know what they're looking for the future State and now we get into the second phase what would you call that second phase.

Mohamed Radwan: [16:32] I would call it if we consider the the discoveries part of the planning so maybe we move on to the design field so the design phase is more like engineering job so you got old this input you got all the requirements from the customer you highlight you your document or your understanding of that requirement and a document.

And you got the sign in from the customer this is very critical as well then you start working on the design so some in some places you correctly work first on the conceptual design.

Old work on the architectural or high level design regardless of the name you need to start thinking about your. Solution from high-level perspective don't go into the bits and pieces of the configuration lion or the timers on you just.

[17:16] Making the I need let's say you start talking about I have this application I need to think about maybe what is the transport protocol shall I use mpls what IP, making some this design decisions and based on that you for each design decision you highlight why you actually went that way so example you have two options shall I go all the way for is is.

So then based on the customer requirements you know customer cares about scalability but he doesn't have an operation team which understand is is so that why you have decided to go with all space so maybe during this phase you would say we'll go for ospf for this and that reason we're using we are going to use multiple areas example and and so on so this is mainly what you need to capture on the high level design and then you start working on the low level design with the configuration template and so on.

Zig Zsiga: [18:08] We jumped right in so we have that design phase and we have the high level design is what we're talking about which is normally a high level design document but then are you also saying there's a low level design document as well.

Mohamed Radwan: [18:21] Maybe now people becoming more agile and they want to reduce this amount of documentation you can see it in some places not everywhere so in some places you have maybe to design document in other situation maybe they combine it and recorded a single design document but regardless I mean as a concept usually you have lawyer.

Conceptual design with your this decision this design decisions and rationale behind it maybe you can put it in the executive or the first part of your design document and then the details will become later and which include all the knobs all the timers all the IP addressing and even the templates about your design some cases you put also the steps, for the implementation Team how to do this installation axis me and so on.

Zig Zsiga: [19:17] That's a lot right so that that's a lot in the design phase so we talked about conceptual design. We talked about HLD the LLD and I like the agile approach I've actually been saying that a lot lately. More recently where people don't want a single HLD and a single LLD, do they want just with merged together in one document and a lot of cases I don't know.

I know I've seen this on my side of thing so it's where we have an HLD done and we have an LLD done and we just put them in the same document and it, it's kind of transitioning now to a more formal hey this is how we do a single design document because that's that originally it was like oh we did the HLD and the LLD and we just popped it into the same word document and we gave it to the customer but it doesn't really tie things together. I mean it's technically still two different documents you're just putting it into a delivery form to one document.

Now things are starting to interweave and it's making a lot more sense because you still have the high level of sections and then you have that low level section in that same section and it's more agile like you said and it's more of a single design.

Mohamed Radwan: [20:23] Yeah or or even the other extreme or which is becoming you know like with DevOps and CI/CD so even some of the customer if you have like an project with an automation we want to automate a network and make it like infrastructure as a code maybe the customer will ask you for the extreme I don't want any documentation just put everything in the Github.

Zig Zsiga: [20:45] Oh yeah put it all together.

Mohamed Radwan: [20:46] Lillian area yes so maybe this will become an or in the following few years.

Zig Zsiga: [20:54] That is an interesting perspective. I wrote some stuff and on GitHub.

Mohamed Radwan: [20:58] Yeah yeah yeah good top of that love you like me.

Zig Zsiga: [21:01] Yeah I wrote some stuff in GitHub. For Cisco live and I actually did like a blog post but I didn't think about putting like the documents and get Hub either that's actually really or get lab sorry and that's really really smart actually.

After the design phase what isn't the next phase?

Mohamed Radwan: [21:19] Usually the next phase is the actual implementation so and here's a key sub phase or stage in the project with the transition because in most of the cases the team doing the implementation is not the same team doing the design so how to make sure that when you transition your design to the implementation team they understand the design they understand why you did that why you meet make these decisions.

[21:50] Also may be the custodian of the network of the operations team also maybe need to be aware of this design approve it you need to consider the criticality of the network so.

How you were going to implement it are you going to implement it on one night is it a big bank or if it's a big Network you will carry it on to a small chunks and then you decide on the maintenance window so even the implementation phase is a big one and usually it's taking maybe sometimes if it's a big project to sticking the longest among all the phases

but it's critical I will not I don't want to talk a lot about it today but because we are focusing on design but it has some places it has its own document you know maybe a big window like Cisco if you have like a lot of template like Network implementation plan Network ready for use which I believe it's a good standard.

To like and it's helped like a hand over between the design team and commission team and operation team to make sure that your design actually worked as expected and the implementation, follow the standard and follow the steps which you highlighted as a the main designer for this project as.

Zig Zsiga: [23:00] Now that that's good I think that just highlighting that what it is and the handoff that's that's critical and I would say like there's probably some critical lessons learned on how you properly hand things off right like like because again you're not implementing it I'm not implementing it my team where I work implements everything not me and so you know we had the planning we have the design phase and then we have some sort of hand off and then I have 30 Technical leaders and engineers and technical folks that are going to go implement this stuff now and so I have to make sure I hand it off and I'm going to show them that would be the same for you now as you hand it off.

Is there any way that they can reach out to have questions and comments concerns?

Mohamed Radwan: [23:43] Yeah usually we prefer to have it like in a meeting your new usually we send them the document and the best way is have it like a session where they come you start presenting your LLD. Start telling them maybe the things which you believe it's in will not be very obvious and you there will be some negotiation and discussion and I believe during my career sometimes I had a very tough discussion was Operation team because for them then they start doing this trade-off so maybe for you as a designer you want to have the network which is very first Converse.

Very Advanced having the latest technology maybe you want to promote something like Signal throttling but maybe the operation teams the opposite do you want something stable something will fisted something the really aware of so then there will be this trade or so as I told you in some during.

[24:41] Previous project sometimes we have like the optimal design but we have to postpone it a bit, because operation team is not ready they are not willing to take the risk so that's why maybe maybe some people will come after five four or six years and say who did this design why they implemented that way so I believe don't judge with the designer you don't know what the situation the design was done five or three years back maybe sigmatropic wasn't popular at that time maybe I was excited wasn't invented it was only Isaac see that's why he chose that so it's that the decision-making it depends on the situation on the time of the he made his choices that regulation and a lot of things so they're always like that why I learned it may be a bit later in my career that I document all my irrational. Behind my design I don't keep it open like that.

Zig Zsiga: [25:35] That's good document all of your decisions right all of your why you made those decisions.

Mohamed Radwan: [25:40] Why am I designing.

Zig Zsiga: [25:41] Five years from now you're not going to remember. I can't remember what I did yesterday what decisions I made so this goes into a kind of you know we're talking about the planning phase a little bit more. I think it would be beneficial here to discuss how you go about identifying your customer requirements as a network designer.

Do you have any specific way of doing that any process I know you mentioned that you have a you have templated questions that you've learned over the years that work but is there like a process you follow.

Mohamed Radwan: [26:17] Yes for me usually I start with the I do a quick research about the customer first thing I go to the website and understand that business understand there may be a vision what is the major major project that you are working on now and some cases even I go to and read the book.

Balance sheet check the finance still see the stock market are they going up or down then see something very critical I understand that million more the key, Personnel within the customer and then I go to the other details if there's an RFP I make sure that I read the RFP carefully and if I have like usually if it's out of P I have like a period of time where I can send my clarification questions. If it's just like a project that it project then I go and meet the customer I as I told usually I have a couple of sessions, today I'm going to meet maybe the top management or ask questions more about the business.

Next day I'm meeting with the engineering team. I asked question about the design and architecture guideline and so on maybe the third day and meeting the people on the ground like the operation team the it system team so I ask more deep technical questions.

[27:36] And then I try to document all what I captured through this field in a document as I told you whether we call it the due diligence the customer requirement document or the data Gathering report yes and usually as a way of disability I want to keep any my requirement and I want to give the customer requirement.

Simple and almost solution I want to link it to a specific requirements example and choosing mpls because the customer was looking for a scalable solution with support segmentation so.

To have something like that to support this decision, how to place Back Your Design This design decision to a specific one or multiple requirements.

Zig Zsiga: [28:25] Now that's perfect right. I always say when I first starting out in design. I always found out that there is implicit items and there's explicit items and you as the designer have to weed through all that and you have to learn what questions to ask and how to ask them and how to get that information from the customer and and today and I think I've talked about it a couple times lately.

I've started realizing that there's just a lot of unstated requirements by customers these days and what I mean by unsteady requirements is that they don't actually come out and say them anymore they don't say that they need reliability they don't say that they need availability they just assume it's going to be available they just assume that's going to be reliable now it's on on us as designers to determine that level of availability and level of reliability of an environment and it's really not.

It's really a shift from the network focused availability to the applications and services availability.

Mohamed Radwan: [29:28] It's a very important Point here. A hypercritical adventure is delightful and important.

Zig Zsiga: [29:33] You're going to start using that now too yeah.

Mohamed Radwan: [29:35] To get to the yeah and even some time maybe the customer doesn't. Good not all the customer IT background so maybe they heard about a tear maybe they are using the wrong word to describe the requirements so that's why you need to be very careful when speaking with your customer, try to make sure of that.

Then sometimes you need to educate the customers didn't yeah maybe is that what you want why you want hundred percent availability that their application really looking for that why you want me to send something yeah so.

Zig Zsiga: [30:10] Hundred percent availability I mean that's going to be really expensive right now.

Mohamed Radwan: [30:13] Yes exactly.

Zig Zsiga: [30:14] There's there's a it may not always be expensive for the availability of looking for but if you're asking for a hundred percent availability that is going to be a very expensive solution.

Mohamed Radwan: [30:25] Hundred percent yeah and why maybe if you are a Telco or maybe you're having like a hospital in critical service also or mobile core yes but if you are maybe a small shop and sitting with pizza so while you're here and you will get collector it depends it depends on the business and yeah.

Zig Zsiga: [30:44] The key word is it depends right it depends so I think there's this kind of goes right into another question I have right so how do you know how do you properly handle the constraints that are given to you as a network designer?

Mohamed Radwan: [31:00] Shoot for me first I need to make sure they are actually constrained so I want to make sure that you should because I speak with the stakeholder and usually within companies maybe there's some political marriage sub maybe there is some logistics or lobbies so someone person maybe he'll tell you about the constraint which would end up is not, through so first I need to validate actually disagreed constraint or not and then if I have multiple multiple constraint usually I try to read them and see how critical they are how how will they affect my solution and based on that sometimes some people you may think about constraint something like a risk.

[31:46] Then you need to start thinking should I accept this constraint shall I mitigate this constraint or should I transfer it or maybe try to convince their customers that maybe they can live without it or something some container I have to accept if it's like more like regulation, or a standard example if I'm working with the hospital I need to follow this referral marketing and you are selling it to follow the socks standard zone so this is constraint is a must I need to follow it but if somebody told you by the way leonie

There's a constraint on budget which is we cannot exceed that amount and this came from the technical team usually it would be better if I go and confirm this such constraint with some somebody like the CFO or the procurement team. Or the opposite if one of the network team told me some constraint about the application,

I may go and double check it from the application owner and if it's real constraint then I try to adapt my design to match this constraint if I want to give this application may be a higher quality of service, or allocate more than words for it or maybe the extreme that I can create a dedicated traffic engineering panel or a VPN for this specific applications.

[33:01] Usually one of my smart manager one of the guys who really affects me a lot and I am still in contact with him he was one of the major I work with in France Telecom usually used to say to us when we started our design career usually technical technicality is not difficult you will find a solution. Any problem from technical perspective you'll find a solution you are good and Technology have a lot of options usually the showstopper usually how to operate it how to build it, some business showstopper that yeah I still I still believe this is almost correct in most of the cases not hundred percent but usually you can find you will find a solution technically.

Zig Zsiga: [33:45] No so I think there is something at the beginning of what you just said that is a big takeaway and I'm trying to emphasize this because of it I'm not going to say that the keyword that we just said and I can say hypercritical, again for the 20th time right this is just a big take away you have to validate the constraints are actually constraints and I give an example like I will go on with a customer I'll go on site or I'll have a call and they'll say yeah we need this it's a requirement and it's a technical person they're saying that they need the technical solution maybe it's maybe it's sub-second failover maybe it's you know fast reroute I don't know whatever highly complex solution and so I'm like okay well why do you need this okay and I will keep asking why until I get a valid response that says why they need it because a lot of times and I hate to say this a lot of times the technical folks they don't know why they actually need it they're just copying configurations scripts or whatever or they weren't there when, the design was created and they don't know the reasons why now there probably is a reason why I will say that to I'll tell everyone that if you are brought into a network and you never been in a network before.

There's probably a reason why everything was done the way it was done in that Network you may not agree with it you may not like it. Right it may not make sense to you right now but there's probably a reason.

Mohamed Radwan: [35:10] Hundred percent and this actually brought something to my mind I believe one of the I took a course recently about strategic thinking thinking or something like that and you're talking about like or negotiations or and they are talking about the five wise so usually in order to as I mentioned to validated you you start asking why I need sub-second conversions why.

Because of this application wise application needs of second requirement because it communicate with the people between East Coast and West Coast what is the distance so you start asking him why why why, it's not the way that attitude is already like real detective or something that just is trying to validate it and without then do you know after the doing all this flew back and forth discussion you know it's actually required 10 millisecond not sub-millisecond that's happening.

Zig Zsiga: [36:01] Yeah and you figure it out right and again and then it's documented and then the company understands right that hey this is and you can reduce that complexity a little right you can mean the configuration of the device is bloated because people don't know why things are done to begin with.

I think this goes into a good question following the discussion right so we make these designs and you and I are designers. We do a lot of design work and our designs are awesome. They don't have any issues, they never fail right so...

I'm being tongue-in-cheek or whatever but you know so what do you do as a network designer when your design fails.

Mohamed Radwan: [36:41] Take a vacation don't go to the office next delivery I'm joking.

Zig Zsiga: [36:45] Yeah I'm sorry guys I am on PTO I am and I am going to the US and I'm gonna go to Hawaii and hang out with Zig.

Mohamed Radwan: [36:56] Zig you this is our challenge usually people you know go ahead blame the network usually if there is any problem it's because of the network then you have you need to, spend some time to convince everybody knows not a little bit because table my design is correct it's a very dark location of the database and so on or the customer.

Yeah so yeah so yeah we faced it and you any all of us I believe most of the energy we have like design failures of the grand challenges or maybe something happened so for me I try to do like wearing my operation on problem analysis hat and we'll start building with this fish bone idea what is the problem what are the possible reason and then, start thinking of taking this one and one of the.

[37:46] Any question I thought asking myself is this net designed feeling for a common use case or it's a corner case because usually you are a very good designer like I like myself Zig and you know usually when you go and do a design you do it for a majority of cases we don't design a solution for, three multiple failures or to failure or three field at the same time usually do we have an ultimate and reasonable budget solution we designed for a single failure or so on so this is one of the first question to ask this failure because normal situation which we are considering our design or maybe this is like disaster or maybe like a power outage for the not to Data Center at the same time.

[38:33] And if you find out that actually it was something wrong in your design it's not the end of the world we know that the design is iterative. Maybe this is not some I'm not trying to be like Defender but we sometimes you miss something or maybe somebody didn't capture all the requirement or somebody give me the room requirement so you design your solution and it wasn't a hundred percent optimal so not a problem you go and do some tweaking for your solution and keep improving it so I believe over the years even for service providers and the big Telco usually each year they have projects like enhancing the network so you need to keep enhancing the routing protocol. Keep enhancing their security so it I mean it's a trade off and design fails it's part of our life and To keep improving.

Zig Zsiga: [39:25] I don't know my designs never fail man they never fail no no no.

Mohamed Radwan: [39:28] Yeah.

Zig Zsiga: [39:29] So I would say there's a couple things and I would add to this right just to be clear sometimes our design fails because we're not given all the information so there's that aspect right we don't know all the requirements we only have a small picture or a part of the picture or the part of the full View and so when we design something we have to keep that in mind that.

You know if they don't give us all the requirements or we don't find them for whatever reason you know the design may fail because of that so there's that aspect we have to tweak that design moved to go in and tweak it or redesign altogether depending on those requirements that we might have missed in some fashion so there's that aspect and then honestly like, sometimes we make mistakes right like it it's ours it just happens right we make mistakes we over judge something I mean it is a design active and once things go into the implementation phase that we talked about the beginning that's where things should get revalidated and those implementation engineer's or Specialists that are you know typing away on the command line there with the devices they should be able to have.

[40:34] A way to reserve the right to come back to the designer and say hey I don't think this is going to work here here's the you know here's my reasoning right here's why because x y and z, and then the designer come back okay I know you're right you know what I did is it going to work here here's here's what you do here or are a better way from that.

Implementation engineer perspective is not to come to the designer and say hey this isn't going to work what do I do it's probably more like hey mr. designer this isn't going to work here's what I suggest we should do is that engineer that implementation engine should have an idea of what would work and then that Designer can be like yeah no that's good let's do that you know.

Mohamed Radwan: [41:13] Hundred percent agree and this helps Also may happen that's why I believe one of the good practice that you involve the implementation and operation team during your design phase so maybe you get some feedback from them so what is the current state of the network what is our utilization what do you think that if we going to run this feature is supported in all the our Hardware or maybe we have a new hardware that doesn't support this feature even if you are a hundred percent aware you will not have all the small details like the custodian of the network or the solution like who are the implementation open operation team so consider them as your eyes and of the network so it will help I agree with you the.

Zig Zsiga: [41:55] Awesome awesome so I have this question now about how do you as a designer work with different vendors.

Mohamed Radwan: [42:05] So here I would assume I'm not a network this designer inside the window our assume I'm more into like a system integrator. Which dealing with multiple vendors and developing some solution for my customers using this winter so I would say this vendors in that case for me like a supplier like a partner, we do Alliance together the success of project it depends on we having a good relationship. With the this vendors so originally from Human perspective I have to keep this good relation, I have to keep meeting them regularly I need to if I can meet them in person or if not I can at least follow there, product pages understand what is the new solution read the data sheet and also for me if I have multiple winter usually accurate like something like a heat map so it's example for when it's come to routers I'm I have been tour ABC Venturi is very strong when it come to maybe on the internet layer when the Ruby is very strong in the core because he has this, first switching console so I thought I keep a distance the same distance and I have to be fair with all the window and also I need to keep.

[43:24] Good religion and be I'll be up to date with the technicality and I believe one of the four companies we used to have like a score sheet for each vendor.

It wasn't only technically it was like, we have like a weighting system maybe we give sixty percent for their technicality maybe 20% for the financial history and financial prediction and maybe another 20% for the customer feedback H it is every and each car company they have their way, but maybe the bad mistake is to maybe to get in 200% was one one vendor I only work with this vendor regardless of the customer requirement regardless of my needs I will always work with that when we have to be smart and and actually choose the right solution and the right vendor based on the customer requirement and and sometimes there will be some regulation, even you want to choose this vendor maybe this guy this can't country you're not allowed to send Ali American device or Solon we face that in some situation where we will were biblical example a multi national network for a big customer and then we there was some good countries for he has some branches there and in this Branch we couldn't use this vendors we have to find another solution so that's why you have to be flexible and you have to be having multiple relations with the vendor with different vendors are you.

Zig Zsiga: [44:53] That's a very interesting take right I think that's good and I think that's beneficial you have to keep the relationships with the vendors because you're going to leverage them and then you have to be able to leverage them to fix or to solve the customer issues appropriately and I think that that that's how we do it right that's a good take I haven't heard that take before I like it thanks Mohamed.

All right so. When we're talking about Network design and design decision specifically to make these design decisions there's always a number of priorities that that are there from the I would say The Human Side you mentioned humans earlier right The Human Side so there's a number of priority so these are not necessarily technology priorities and implementation priorities are more human priorities how do we or how do you manage those competing priorities when making their Network design decision.

Mohamed Radwan: [45:47] Yeah by human priorities the zig you're talking about maybe like from time perspective from a physics perspective or also from Technologies everything here, so usually for me when I have this contradicted contradicting priorities or maybe different opinions and so on I thought find like I want to make my decision and I want to be fair.

[46:11] So I thought create something like waiting system and I start thinking about some criteria so as example I have like the customer he will he want to do this to Lucien as fast as possible but the other team they want to do it with within budget or so on so I start doing some waiting, good idea from technicalities solution a is will cost you this amount from scalability from so on and I start giving maybe some wait for each of these criteria and for each solution I give a fair score so this solution it will take eight out of ten from technicalities or option A or option b will take 7 and and this don't that I try to make like a more objective weighting system and I can choose, which option makes sense, when it's come to maybe other priorities which is cannot cannot be evaluated I start understanding from my circle talk to the customer and understand the their Community understand the reason behind them and I can see sometimes if it's a must I try to be flexible and this Purity is required Maybe, adding more resources to the project we for a small period of time to be able to.

[47:36] Finish maybe if I have to two phases and these two phases become very Purity and the customer want us to finish them maybe before the end of the year, we can wish it was the customer okay mr. customer I can achieve these two priorities in.

This time but maybe we need to agree on the CR or near you need to agree starting next year will extend the project so on so there is a lot of negotiation skills here and a lot of give-and-take between you.

Internally and with your internal customer and between you and the customer as well.

Zig Zsiga: [48:13] So I like that you put that last part there that there's a lot of give and take that's that's what I've learned at least is there's a lot of give-and-take and being overly communicating back and forth. Have that dialogue with the customer and then I like to have some quick wins with the customer. I like to hit something that they care about, as soon as possible and then that way that builds that relationship so I'm going to say that again one second there was some extra noise. Figure that thought process my head.

He corny or not corny the marketing trusted trusted advisor trusted leader term but like literally we're building trust like I will okay we're going to do this the literal by this date at this time and we're going to have it done and that will be the first deliverable and we knocked that one down we get that date done we don't slide that date ever and then and they might have a priority of hey I need this for a budget situation like we are doing budget. And our budget will be defined on this date so I need that document that you're writing to you know it might be a future plan document for the next five years I need that a week before that day so that I can properly put those things in my budget plan and get the budget for all those items that I need and that's where this sits is that's one of those priorities that you want to do.

Help them.

And build that trust so that in the future you can like okay well we have that trust now and we can kind of manage those competing priorities we can't do everything all at once it's not possible.

Mohamed Radwan: [50:06] A hundred percent like the concept of quick win is and building The credibility building the rapport with the customer because actually even for our if you are like a vendor and dealing with the customer or supplier or service provider even if. Yeah I need to achieve this priority you will do some investment from your side or maybe you’ll have some peace some money or so on or additional resources from a by any means I believe you you will repay it's a customer will really appreciate what you did for him, and maybe he'll give you another project or we its it will you will get that repaid for sure.

Zig Zsiga: [50:45] Well it's circular right like you're building that relationship and it's just a constant like you keep providing value and if you keep providing value I mean it's going to be clear that hey that customer is going to want you back.

Mohamed Radwan: [50:56] Yeah yeah.

Zig Zsiga: [50:58] Especially we're talking more of a consultant all right that's.

Mohamed Radwan: [51:01] Yeah.

Zig Zsiga: [51:02] The customer is going to watch it back like you're literally building them what they need and you're making that into making their job easier right you're doing this work for them and they can take what you had what you're given them.

And they immediately apply it to everything they're doing all right.

Mohamed Radwan: [51:20] Because we are human at the end then yeah I was going to say just in one minute that we are human too people did usually they will buy your solution or by your service because they like the solution for sure but also they need to like you if they hate you maybe your chance would be difficult so.

You have to build the trust you have to build the credibility Builder relation itself will help you as a designer as an architect.

Communication skills and interpersonal skills are very critical for a designer and architect.

Zig Zsiga: [51:52] There there you have it that that was a quick that was a good take away right it's not just the technical stuff we have to have those soft skills communication we have to be able to build those trust the trust and that'll help you, in this role as a network designer all right so let's talk about those next question and we'll get into more of a maybe certification Focus items but this is about budget and I would like to know your perspective on how budget affects you as a network designer your role as you go through making these decisions and Designs how does a budget.

Mohamed Radwan: [52:23] Let me ask you a question Zig, how many projectss in your life haven't had a budget?

Zig Zsiga: [52:32] Pretty much all of them right now my budget might be pretty high right I might you know the customer might come to me like yeah we have five billion dollars for this, this massive idea and in reality they don't have five billion dollars you know if they have five million dollars and it's still a lot of money but five million dollars isn't going to get you that five billion dollar you know idea in your head.

Mohamed Radwan: [52:54] Maybe.

Zig Zsiga: [52:54] So so you know that's a great question thanks for asking it for me every project I've been on has had some sort of Budget.

Mohamed Radwan: [53:05] Yeah I agree so what I want to achieve that usually 99% only five nines of the project you're always have like a budget constraint there is a finite resource nobody will tell you you have an open, budget for this project to go and buy whatever you want so maybe you go and buy Cisco see arrest to put it like put it on.

For the entrance of your building so no at actually usually there is a budget and usually you need to be aware of.

The customer budget and the spending in the to use solution wisely for meat was sometimes a bit tough because in some situations in the special in the big companies for you as a designer you aren't actually aware of it.

Let's go score the commercial component of the solution so thats why sometimes you start very unaware of.

[54:00] Hardware but you're not aware of that discount percentage or not maybe a hundred percent aware of the labor cost or so on so that's why it's a bit challenging for some of the designers to build the solution within budget but this something you will learn over the time and actually depends any this is Case by case and Depends for each company for each.

Countries or regions as example if you're working maybe I want to give a bad example that if you're looking to be in Africa Where I Belong it's completely different than working having Solutions in the Gulf area or, or here in Australia because the in Africa as you mentioned maybe $100,000 is a big project for them so usually you have to be very keen sometimes will become very difficult to choose an optimal.

Vendor optimal solution because this vendor will be a bit expensive so usually you have to go maybe for the second or the third option, so that's why yeah we work in this circle and tensors consensus is becoming even more difficult for the designer to get to get the best movie in many cases they will tell you I don't know that resilience is important but I'm willing to take the risk.

In some of these situation you will have proven Build Your solution with a lot of maybe single router single link and so on so.

[55:21] Others as I said budget is always there in most of the solution just the kick of it and what you need to take away from this question is budget is there there is always the finite resources and.

[55:36] You need to be aware of the where the solution you are implementing as a designer if you can get some insight from the sales team about the budget and about, the solution breakdown so it can help you to define the right solution.\

Zig Zsiga: [55:52] So that that is great at a couple things that I was thinking about here how I position this so there's there's certain circumstances with budget where,

I've had these certain these situations over the last 20 years where a customer says hey we're going to do this we have this huge engagement we need a ton of stuff you know equipment and services and you know they just have a lot of things that they need and so we start talking and I'm like okay but you have a budget for this right now the yeah we were budget we're good.

And so we don't talk specifics and then you know we start designing a solution and then we get to the end of it all and we give them a list of materials right that bill of materials which has all the devices on it and all the hardware and all the software and all the licenses everything and you know.

So we so we get there right with this list we give this list to them and they're like in and you know the prices like I don't know maybe it's a million dollars let's just say right or maybe it's yeah I mean a million dollars and they come back to us and like oh our budgets only $50,000.

[56:53] Like okay we are not on the same page there so that situation that's a bad situation to be and we go well we had be trust for the would be truthful we had to be upfront and say hey well you know you're asking for something that is a lot more money.

And you only have 50 thousand dollars so we got to start pairing that back down and get you to where you can be but again it's not it's not we're talking about two different things talk about apples and oranges right we're talking about two totally different ideas and then the other thing is that when we have these situations where there is a budget constraint because there's always going to be a budget constraint I like to let the customer the side I like to.

Provide them all the information I can. I'll use a layer 3 access design right so layer 3 access design means that.

[57:37] For those who don't know you have your switches your local area network switches and instead of running layer 2 like spanning tree between all your access switches to your distribution switches you'd run layer 3 and so you kind of cut out all of those you know spanning tree loops and all that kind of stuff and you rely on layer 3 routing.

I would in some of those circumstances for Designing that solution I talked to the customer like hey this would be a really good solution for you, it hits your requirements perfectly.

But there is a constraint here of budget right I need to be very clear you know the license to do routing on these local area network switches.

Is a lot of money and you may have 400, local area network switches are access switches and you have to buy 400 licenses and let's just say they're ten thousand dollars a license I mean you can do that math adds up so it comes down to you know this is what you could get if you spend this money.

Or we can do a different design that you don't get layer 3 routing and you have spanning tree and you have you know links that aren't being leveraged potentially your you know redundant links that are blocking state or you have poor channels or whatever but again there those are the.

[58:51] Compromises right budget is a big deal but we have to present it to the customer with the present it to whoever we're talking to and show them so that they have that Avenue they not Avenue that ammo when they're talking to their leadership say hey we need an extra $10,000 times 400 right like you know we need extra money so that we can do this and we can ensure whatever level availability and optimization excetera.

Mohamed Radwan: [59:18] I agree with you completely Zig and you this is also again and he coming too building the relation with the customer trust so if you don't have the trust the customer will not trust you so maybe its customers say oh this guy's actually just want to oversell this at the extract work equipment to us to they would say no but if there is a trust and the understand and you explain to them you were very open with them they would be your Advocate internally for your wedding their CEO within their procurement and I believe in most of the cases you will get what you want but this for them so but we need to work on this relationship right there.

Zig Zsiga: [59:56] All right so certifications man certification so what is the most important certification in your role today as a network designer.

Mohamed Radwan: [1:00:05] Yeah there are some obvious certifications that let me start with an obvious.

Yeah so I believe yeah maybe nobody will stop thinking about CCDE and so on but for me I found some other certificate very useful, example TOGAF which is more like Architects but is very helpful for the as a designer I still remember when I started working as a designer 2005.

My company told me you have to take PMP. I don't know why, I'm an architect. We know it will benefit you and actually did so understanding the project and understanding the time management people management constraints so this will help you at least when you start putting your solution together.

The additional elements of the distribution, the people element of it and so on, I would also suggest something maybe I did.

[1:00:55] It's good because in some situations in one of the previous roles I have as an architect one of my main roles is not only to design the solution but to design how to operate And build a solution so I was dealing a lot with the different team and then I need to understand how what is the service desk will do what is the, position teams the problem management and all this so this will not be an old cases but I'm just trying to say in addition to the.

The normal stuff which everybody will tell you studies to work on your foundation maybe CCDE, CCIE, CCDP and so on but also there is some additional certificate even CISSP.

CISSP I found it very useful because it cover security from different aspects and security is very important for you as a designer to how to make sure that your solution is secure.

[1:01:51] Truth is sometimes we don't go that deep for physical security or Zone but I was involved in one of the SOC security operations center design we will build it as soak for multi-tenant so can you how can you imagine how to cure this one should be because actually our manage the security for multiple customers so the amount of security.

Workshop we had to this to decide even how to have a short physical security how to ensure network security how to ensure that the customer will not share any data and so we built a lot of multiple layers so and I believe this kind of certificate will actually add to you and will keep your unique maybe a designer or a better version of the network design.

Zig Zsiga: [1:02:40] That's a great that was a great.

Mohamed Radwan: [1:02:41] What else have I missed I have I missed something exact you found useful as well.

Zig Zsiga: [1:02:46] The only other thing would be like like business architecture I would mention like.

Mohamed Radwan: [1:02:50] Yeah yeah exactly that's bridging yeah.

Zig Zsiga: [1:02:52] And I think that gives a whole different perspective on, how to use it again we're talking about something that is critical to Bridging the Gap between the technology thought and the business side and togaf is great and I tells green I use both those. PMP is great as well and obviously we're both CCDE we're going to tote that right we're going to highlight the CCDE as well so we already know that.

But yeah I think I think a right business architecture is the other item I would add to that list because and again that's not a certification per se it's training and but it's it's very valuable it really does help you with the business side of things.

I'm so the last thing here you know where can everyone find you on the interwebs I like to ask that question you know I want people to be able to contact you keep the conversation going so once again we're can everyone find you on the interwebs.

Mohamed Radwan: [1:03:40] Yes I have my website it's agileis.com.au,

I have also my YouTube channel it's Muhammad Rod one.

And you can find me on LinkedIn the same name Mohamed Radwan as well and Twitter are not really active on Twitter that. I'm starting. Zig is actually one of the guys who convinced me to start using Twitter. MohamedabRadwan on Twitter as well.

Zig Zsiga: [1:04:09] And I'll have all those links in the show notes just to make it really easy they'll be a little section will be the Muhammad read one section was a how to keep connecting in our keep the conversation going with Muhammad you have all his links and also you have all the links to all that's course content Muhammad has a whole bunch of course content study plan for the CCDE, free Simulator for this you CCDE as well so if you're studying for you know Network design or well for something for the CCDE or if you're learning more about Network design there's a number of things that you can go and check on Mohammad site and I'll have all those in the links just so you have them that I miss anything there, on that part of the only I did I think I got it all the next thing I want to say is that Mohamed also has an e-book it is the it is titled the CCDE The Practical Guide.

Mohamed Radwan: [1:04:58] Yeah this is CCDE The Practical guide.

Zig Zsiga: [1:04:59] All right cool and it and we're going to do a giveaway. Just so everyone is aware we're going to do a give away so I'm either going to buy one of his books or he's gonna give me one for free to give away one of those situations and we're going to do a giveaway I don't know how long it'll be but if you want to be part of that giveaway you can go to Zigbits.tech/MR.

Mr and that's just for Mohamed Radwan shortlink. Zigbits.tech/MR and you'll get all the details on how to be a part of the giveaway and then we'll do it for a few weeks and then we'll just give the book away.

Hey Mohamed, I appreciate you buddy thanks for joining today. I'm going to close out this show and we're going to wrap it up man.

Mohamed Radwan: [1:05:33] Brilliant and thank you everyone and thanks for having me and have a good day.

Zig Zsiga: [1:05:38] Have a good day man alright hey friends nerds geeks and ziglets that's going to close out today's episode of the zigbits network design podcast.

Where we demystified the role of the network designer remember at the beginning I said that you have to find something that you love and life that's what you want to do you want to be happy so hopefully this helps you demystify this role hopefully you understand what the designer role is a little bit better today and if you want to add some comments or see any information that we talked about.

You can join the show notes. That's where we have all the links for Mohamed and all the content we cover today will be on that show notes page.