Dear friends, brave critics, quiet readers, and secret admirers, SDSM is coming to an end.
I can't claim that over 7 years I have covered every topic in networking or that I fully explored even one of them. But that was not the goal. The purpose of this series of articles was to gently guide a young student into this world, taking them step by step through the main gallery, providing a general understanding, and protecting them from painful wanderings in the dark corners of the mind of Olifer and Olifer in their agonizing attempts to find answers on how to apply all this in real life.
SDSM was planned as a short practical course 'how to learn networking in a month', but it turned into 16 (actually 19) long releases, which we even renamed 'Networking for the Toughest'. The total number of characters exceeded 1,000,000.
It would be reasonable to stop at BGP, but diving into MPLS without IP doesn’t quite work — it had to be seized. Perhaps it would have been better not to tackle Traffic Engineering, but once I got into L2VPN, how could I stop? Hardware architecture is an inseparable preface to QoS. And QoS has been so long requested that I couldn't help but write about it. There’s nothing to discard at all.
The last article was supposed to be a collection of best practices (suggest a pleasant-sounding synonym — and I’ll replace this calque) for the design of provider networks, but over time and with experience, it became clear that this is not only a vast array of approaches but also fertile ground for verbal sparring. And why should we stop at providers? What about telecom operators? What about data centers? What about corporate networks?
You can't say: do it this way and it's right. You can't teach an engineer to design — they must grow into it themselves, navigating through their thorny bushes.
This is precisely what SDSM offers — a barely noticeable path from simple to complex.
Look at how much has been accomplished...
Many people contributed, some with their hands, others with their heads, to the writing of these articles:
Participants…
- Max aka gluck — co-author of the first articles and author of the 4th part about STP and the IP SLA section in the 8th. Additionally, a project admin for over 5 years.
- Natasha Samoylenko — additional materials, tasks and their solutions for many issues. Also, invaluable support.
- Dmitry aka JDima — critic and proofreader.
- Alex Klipper — critic and proofreader.
- Dmitry Figol — critic and proofreader.
- Marat Babayan aka botmoglotx — author of the issues on EVPN and photographs for issue 14.
- Andrey Glazkov aka glazgoo — critic and proofreader.
- Alexander Klimenko aka volk — critic and proofreader.
- Alexander Fatin — critic and proofreader.
- Alexey Krotov — critic and proofreader.
- Team linkmeup — proofreading the materials.
- Anton Klochkow — organizer. Thanks to him, the project has a lab environment, a broadcasting server, and now its own podcast hosting.
- Anton Avtushko — website developer, who has faithfully served for 6 years. Livestream has long been deprecated, no plugin is supported anymore, yet the site still lives. And lookmeup.linkmeup.ru, stillborn but with a good idea.
- Timofey Kulin — admin and website developer.
- Nikita Astashenko — website developer.
- Nina Dolgopolkova — illustrator. Logo and illustrations for issues 9 and 10.
- Pavel Silkin — illustrator (0th and 1st issue).
- Anastasia Metzler — illustrator (11th issue).
- Daria Kormanova — illustrator (12th issue).
- Artyom Chernobay — illustrator (13th, 14th, and 15th issues and this concluding article).
The article about QoS has become the last in the series. By the time it was completed, it became clear how simplistic and incomplete the earlier issues were. In fact, everything up to BGP is quite poor.
Moreover, readers often find mistakes themselves and suggest corrections.
The idea of moving all of this to Gitbook, brought up by Natasha Samoylenko, seemed so appealing that we went ahead and did it:
Today, a large portion of the articles are available in an updated format there.
Anyone interested can fork the project, make changes, and submit a Pull Request to the master branch. Once I approve it, the changes will appear in the Gitbook.
For now, I believe there won't be a printed book on SDM. I'm not ready to dedicate time to rewriting the initial articles to create a coherent, beautiful, and, most importantly, comprehensive material on networks. After all, there are many interesting things in this life, and I'll deal with perfectionism somehow.
A significant factor in completing the cycle and shifting interests is the change of job.
And a short announcement: our hands are not for idleness, but for writing. Expect a new series of articles about automation.
Source: habr.com
