[cs615asa] Meetup Report
yli142
yli142 at stevens.edu
Thu May 4 01:21:22 EDT 2017
Hi,
I attended google/facebook’s Site Reliability talk last week. The reason
why I chosen it is that google and Facebook are big company and it will
be a great challenge to manage and monitor the backend service.
There are three topics:
Firstly, the structure of data center network based on Clos topology.
The old structure is that networks started to use the "fat tree" model
of connectivity using the core - distribution - access architecture. In
order to prevent oversubscription, the link speeds got progressively
higher as you reached the core. However, the Clos network now manifests
itself in the way that the switches are interconnected. Now data center
networks are comprised of top-of-rack switches and core switches. The
top of rack (ToR) switches are the leaf switches and they are attached
to the core switches which represent the spine. The leaf switches are
not connected to each other and spine switches only connect to the leaf
switches (or an upstream core device). In this Spine-Leaf architecture,
the number of uplinks from the leaf switch equals the number of spine
switches. Similarly, the number of downlinks from the spike equal the
number of leaf switches. The total number of connections is the number
of leaf switches multiplied by the number of spine switches.
Second, the kernel testing. The speaker mainly focused on performance
regression, stability issues and security problem when you test the
kernel. A good way to test kernel is test different part of the kernel
as different models. Because it is easy to review you code, control the
scale of test, and use particular tools to test different models.
Finally, a beautiful women talk about dependency traps and gotchas. At
the beginning, she talked about Micro service, load balance and vertical
scaling. All of these technique are popular. The first time I try to
learn such field knowledge is to use NGINX as my server container. It
will light and efficient. However, when you work in a big company,
stability of your service is very important. And the number of path for
resource will be large. Thus, we need a ACL(access control list) , s an
optional layer of security for your VPC that acts as a firewall for
controlling traffic in and out of one or more subnets, to help us. It
will make you service more secure and stable . Another interesting topic
is data dependency lock encryption. Actually I don’t quietly understand
it. It said two server database both have the key. If one crash and
reboot, it could get key from another. However, if both of them crash at
the same time. What should be down?
Thanks.
somre reference:
http://www.networkworld.com/article/2226122/cisco-subnet/clos-networks--what-s-old-is-new-again.html
https://en.wikipedia.org/wiki/Access_control_list
More information about the cs615asa
mailing list