Launched: East African Network Operators Group - EANOG

There was no noise, no pomp, no beer :-) just me and a few guys discussing what we felt group dynamics would help with. Our roots might not yet be firmly in the ground but the elements are definately cooperating with us.

Some background: I was an idiot in the past. I broke more networks than I fixed (don't assume I don't anymore), I steadilly improved sometimes in different areas at the same time. I read, labbed, listened, watched, failed,passed, tcl'ed,perl'ed,cabled till my eyes teared, my fingers moaned but I carried on. Now I can even classify it as fun.

Computer/Data Networking as a career has been very rewarding and along the way I picked up quite a number of good friends and memories.  Im looking for more. more 'networking', more collaboration, more learning. Won't stop till I get enough !

It is not always easy to get information in this industry in a timely manner. But i know with certainty individuals in the industry know alot more than they let on. I also know that peer led training is more effective than anything your HR can ever manage. Plus wouldn't you wish for more meetups you have control of?

I also know that we are in a growing economy and ensuring any one that calls themselves a networker in Kenya needs to up and modernize their game for the future is really up to us. Some of us are more exposed to others, some are 'sharper' - not me:-), some have access to 'hidden toys'...ahh...life

Thats the bit I want to play in. I want to see proper policies, new protocols or contributions to them, white papers from our market written by us, I want RFC's with my name on them for all the fame and glory that brings (none), I want books I want all a lowly third world networking nerd can have before the bits (apparently my brain is a network)overpower him.

Yep someone needs to watch my Effing back !

 So its great to have the mailing list up at http://orion.my.co.ke/cgi-bin/mailman/listinfo/eanog. We'll grow as we age. There is a huge need for an online presence and we'll work on that, for now Im just happy the meeting happened....Special thanks to the folks at Ihub and everyone that came and or offered support and advice. Same goes for Riyaz for the last minute push to have this started - Heck we had a guy from NASA - We'll be seeing you around when Im not on the moon:-) thank you!

East African Network Operators Group - EANOG

For a long time I have wanted to have meetings where the dominant discussions are around networking and networkers in general.

The main objective would be educational, but at the same time would give vendors a chance to talk to engineers about newer technology they are playing with - like openflow, vxlan,nvgre and products on the sidelines.

Engineers would get to talk and share their experiences.

So today we fired the first shot in the dark. An invite to the IHUB (directly opposite uchumi ngong road) on saturday the 19th at 2pm. If we get even 10 people interested, this ball will start rolling.

There will be obvious issues along the way. Top off my head is how much time can I/we (Riyaz and I) commit to this endeavor in the beginning before the group gains traction?

How often do we meet? what do we discuss? (I expect what to discuss wont be an issue). What if we dont get sponsors? do we even need those? can we put together a community lab? community wi-fi? are we doing ok as networkers? how many ccie's are in Kenya? etc etc....see you there....

So in a nutshell we're forming an educational collaborative forum for the coordination and dissemination of technical information related to networking.

We'll leverage alot of what AFNOG is doing with more frequent meetings.

Everything from broadband, data centers, security, mobile networks, IP, openflow,wi-fi and anything under the moon we deem important will be open to discussion.

Third world networkers view on Openstack/Openflow and Cisco

Ever since Cisco announced their support for  openstack and software defined networking; I was happy, happy that they won't be left out and happy that maybe just maybe this information would be digested by a company I am 'more invested in'. I cant wait for instance for the 'promised' support on the nexus.

But cracks are appearing. The more the network gets closer to getting virtualized, I realize how silly it is of me to 'wait' and see instead of maybe getting in there and contributing something. Let me explain:

What this will mean is new books have to be written. First to explain the concept, the technologies and in the future its support. looking at the old ones, updates will probably be written almost daily.

And what happens to companies that invest alot of cash on lets say FCoE only to realize customers are not interested, or realize it wont scale, or wont matter squat in multivendor environments because other vendors have decided to go their way?

The data center is still morphing. We still don't have important standards for lets say 'cloud' 2 'cloud' mobility. Again this have to be done. Heck down here using cloud is still hindered by access (say what you want 3G is not good enough, and broadband penetration is still sucky). So while some big companies benefit from companies like Safaricom shelling out some serious cash for several nexus, and going by how accounting tends to be done, expect it to be out of reach for most people.

source: Juniper Networks

So over and above everything else -and there's quite alot to be done - Im looking at how I can make direct contributions to openstack and a few other areas of interest to me even if its just simply learning and talking about it:

We have huge gaps in Africa. It is very easy for technology to pass us by because:


1: the best discussions and forums and meetups happen in Europe or the in the US. When nice things happen around here, they almost always happen in SA, Kenya or Egypt (not much anymore). but travel within Africa is crazy expensive. A flight to senegal from kenya is costlier than a flight to Berlin for instance. However its harder getting the visa to visit berlin - its a friggin pain.
2: Most companies here consume products/technology and rarely innovate. So you won't for  instance find a Kenyan writing RFC's because - well Im not sure why but personally the fact that I dont really get down to work with new technologies enough to be the 'go to guy' makes me feel like I am better off just waiting - until now:-)
3: Groups like afnog are like nanog focused on training.
4: Money and general social security - I have learnt first hand how difficult it is to focus on 'new' stuff while constantly worrying about how to make basic ends meet.  Im stuck in this weird 'fear' loop at the moment thats driving me nuts. but more on this later and probably elsewhere.
5: Its really really hard to focus on a single area of interest (ok maybe Im speaking for me on this one). After 10+ years in service providers, getting excited in data center technologies was not easy - fun but not straight forward- upon realizing i could pick it up faster than most helped.

So much that Im currently looking at HP's solutions for this area (between trying to get the cisco DC certification) and going Hmmm! They (HP) have a full inhouse stack that should make them very competitive in this market in direct competition to Cisco UCS.

I imagine cisco and Juniper at a classic 'innovators dilemma'. If they dont deal with this, companies like Nicira will have a really nice and easy time. It will take time to get market share but I can see them getting there if the incumbents dont innovate. well they will I suspect outright buy a startup out....we'll see...

Third world networkers view on simplicity

Greetings bloglings!

Simplicity in design is often overlooked. We have done some really impressive networks over the last couple of years. The data center networks have evolved. VPN's are now IAAS and Im pretty sure marketing guys will keep churning buzzwords to confuse (and unlock customer wallets).

RFC 3439 states in part:

'The Simplicity Principle, which was perhaps first articulated by Mike
O'Dell, former Chief Architect at UUNET, states that complexity is
the primary mechanism which impedes efficient scaling, and as a
result is the primary driver of increases in both capital
expenditures (CAPEX) and operational expenditures (OPEX). The
implication for carrier IP networks then, is that to be successful we
must drive our architectures and designs toward the simplest possible
solutions.'

there Ive saved you money.

Every once in a while after a design team is done and implemetation happens. Things start going wrong. a link here, a backdoor over there. statements like Oh I dont understand IS-IS can I use OSPF for this segment? are heard!

recently/actually currently Im on leave for a month and it just so happens that a friend wanted me to help them re-design their network. Its a hospital (large hospital) big budget for gear, sadly not enough for 'people' to run/support the network.

Anyway. thinking about it, I dont see for instance why hospitals in Kenya cant come together like banks and do a common core. It would save them al lots of money. If a service provider like Safaricom were to be bold enough, nothing stops them from selling their core to the same hospitals. It would save everyone money and some serious future support issues (considering they dont want to hire people).

the task of building a large scale packet network is not easy it can be done, but its not as easy as these guys tend to think. ok back to the drawings....

Third world networkers view on relaxing

its a rare month. I somehow had excess vacation/leave days and managed to take more than half the number in one go. How long - All of Otober andpart of November:-). My plans are simple:

1: Reverse ccie laziness - As I type this, my sitting room is a mess of bike parts freshly imported to give it a fresh feel. I fully intend to add a minimum 500Km within the month on it.So if you see a nerd on a bike at 'kapchorua' please wave him on.
2: Farmken finally has a product, customer development is at advanced stages. Not a single negative feedback so far. But we have issues: packaging for retail consumers needs a rework. A kilo of fruits is around 15-20 fruits (don't get conned). sell by the dozen, use egg trays, wrap them up with a nice clear cling type paper with our logo - this also preserves the fruit? we'll see....

• The site will feature an online shop from next week - make it easier to cut out the middle men that come to the farm and make it more expensive to buy fruit for retailers.
• We will work directly with some of the larger 'supermarkets' again in the hopes that our fruits reach the consumer faster and cheaper - need to sort out the packaging issue above. How to do that without charging more is the fun question right now. Im thinking we can swing 170 (Ksh) for 30 fruits which is around 1.5Kg.

3: I was at the IGF and sat in two panels:

1. 63. SWOT analysis of the impact of Mobile Internet on Internet Governance in Africa
2. 165. Understanding IPv6 Deployment and Transition  

I had a lot of fun and intend to build up on the two. I also met some incredible people and can probably see a pivot to policy discussions when the tech ta;ks get a bit much.!:-) 

For IPv6, Im working on some new training materials and validating lab output for the said slides. I'll be teaching IPv6 and advanced network related classes. I'll let you know where/when and how that works with my current job - Just FYI, most of the training is internal but hopefully - considering this are free, I'll check to see if we can include outsiders. I think that would be cool.

Im also looking at how we can bring together the network user  group together more often. throw in some comments if you have any idea. I can probably help arrange a meeting venue, maybe some snacks or whatever if we can raise enough quorum to make it worth the time. We can pick on topics to discuss from trends in the local environment, some design issues you have at work, IPv6 whatever....or just catch up. East African Network User Group (EANUG) anyone?

4: Catch up on technology. The data center excites me. So expect stuff around vxlan/nvgre etc etc etc.....

5: I'll relax alot, read - I paid for a full years subscription of Ivan's webinars that I have not really listened to. I expect alot from there....so there..thats my month...oh I'll probably also blog alot....