Efficient Monitoring for Large Server Farms (with Chef and Smule Grapher)
Agenda:
19:00 – 19:40 – Efficient Monitoring for Large Server Farms (with Chef and Smule Grapher), Asen Lazarov
19:40 – 20:00 – Q&A
20:00 – Networking (Talk, Chat, Pizza)
About the Talk:
Monitoring thousands of servers across multiple data centers/zones can be a challenge.
Manually adding or removing servers to the monitoring system is no longer an option, tools like Nagios and Cacti no longer work at this scale. More modern tools usually require a more complex setup in terms of separate time series database, graphing UI but also agents or probes to put the data into the database.
Asen Lazarov, Server/DevOps Engineer at SMule will present his approach and experience with Chef for server provisioning and configuration. He will also talk about Smule Grapher (open source) tool used for all-in-one monitoring.
More about Asen:
I saw code for the first time when I was about 10 years old – trying to type the printed copy of the source code of a game written in BASIC on a Pravets 82 without having any idea what I am typing. A few years later I was able to get myself infinite “lives” in games like Karateka and Prince of Persia which ended up more interesting than actually playing the games. Have been playing with code ever since then which is more than 30 years now.
Over the years I have had experience with many languages and platforms – ranging from debugging crashes in assembly, writing C and C++ code, writing apps for Java and .NET platforms, writing scripts in Ruby and Python and then Html+JavaScript client apps.
Last 8 years I’ve been working for SMule as a DevOps engineer where I joined a mobile apps startup with 20 employees and about a million downloads and have seen a growth to more than 160 employees and more than 150 million downloads.
