About the Author

Gustavo A. A. Santana, CCIE No. 8806, is a Cisco Technical Solutions Architect working in enterprise and service provider data center projects that require a greater integration among multiple technology areas such as networking, application optimization, storage, and servers.

With more than 15 years of experience in the data center industry, Gustavo has led and coordinated a team of specialized Cisco engineers in Brazil. A true believer of education as a technology catalyst, he has also dedicated himself to the technical development of many IT professionals from customer, partner, and strategic alliance organizations.

In addition to holding three CCIE certifications (Routing & Switching, Storage Networking, and Data Center), Gustavo is also a VMware Certified Professional (VCP) and an SNIA Certified Storage Networking Expert (SCSN-E). A frequent speaker at Cisco and data center industry events, he holds a degree in computer engineering from Instituto Tecnológico de Aeronáutica (ITA-Brazil) and an M.B.A. in strategic IT management from Fundação Getúlio Vargas (FGV-Brazil).

10 comments on “About the Author

    • That is a great question, Mark. It all depends on your previous experience.
      If you already have a good networking background, I would say no. I have written the book as a first step into DC technologies for network (CCNA-level, at least), storage, or server professionals with basic knowledge.
      I really hope it helps you!
      Best regards,

  1. Hi Gustavo, your book is amazing really a great contribution! thanks a lot!!

    I am working on a particular issue with “trombone flows” between two data centers and found a recommendation in your book to deploy a pair of LB/FW in each data center so as to have the network services always active on each site and prevent packets being unnecessarily sent between data centers.

    Question: As deploying two (02) physical appliances (FW/LB) on each data center may represent a major challenge/investment (today we have six ASA 5585-X and six ACE Load Balancer Contexs) is it possible to have two instances/contexts of each network service (FW/LB) rather than physical devices on each data center to meet the same requirement? if so is there any dependency/drawback you foresee on this approach?



    • Hi, Manuel. Thanks for your feedback! It´s very encouraging to know that the book has helped you!
      About your question: I understand that deploying a pair of physical appliances can be very challenging for some data centers. Therefore, before taking any decision, I highly recommend that you have the following reflection: “Is tromboning too bad for my DCs?”
      I know some customers that have plenty of bandwidth and very low latency between sites (they mainly use DWDM/Dark Fiber for DCI) and they don´t feel that the zig-zag of packets is really a problem. If that is your case, maybe you can deploy a single device on each DC without worrying too much.
      Nevertheless, if tromboning is an actual problem for you, I think that the pair per site, is the recommended way to go. Unfortunately, I do not believe that your idea of using contexts would change such behavior in the case of device failures.
      For example, if DC1 has an device with contexts A(active) and B´(standby) and DC2 has A´(standby) and B(active), imagining that the device on DC1 failed, you would have tromboning anyway, right?
      So, allow me to put the question back to you: can your DC handle tromboning if there is a failure on the device?
      Best regards and keep in touch!

  2. I have recently bought your book on Data Center Virtualization FUndamentals. I am an instructor in university. are there slides which I can use or figures of the book that I may prepare my own slides. Thank you

  3. Dear Gustava,
    I am reading your book still on chapter 3, i am a very big fan of Radia Perlman :D. There is a particular example [p71, figure 3-18]: ” Switch C with “direct link failure aka signal loss” to Root bridge would not wait for 20 seconds MAX-AGE timer to expire, in your book it waits. Shouldn;t the total time of re-convergence be 30 seconds instead of 50?

    • Hello, kikiAnichka. How are you?
      I’m in Team Radia as well! It’s a shame TRILL is not catching up as it should…
      And thanks for pointing that out. You’re right.
      Taking a look at my original notes from exactly three years ago, I have noticed that I originally wanted to express worst case scenario where Switch C did NOT detect signal loss. In that version, I wanted to point out what happened when the switch needed to wait for the BPDU aging in e1/3.
      However, in later revisions, I’ve changed my mind on the explained scenario but did not change the text accordingly. And somehow it slipped through my reviews…
      I will surely include this point in the book errata.
      Thanks again,

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s