Some pre-holidays stuff I'm reading

Here's a random collection of this and that:

• SANS Launches NetWars CyberCity to Train Cyber Warriors for Defense. I'm not sure why they feel like they need a physical city in which to train for virtual attacks, but they give their reasoning in the article, including:

  NetWars CyberCity participants, which include cyber warriors from the Department of Defense and other defenders within the U.S. Government, will be tasked with protecting the city's critical infrastructure and systems as they come under attack. Cyber warriors will be presented with potential real-world attacks; their job is to defend against them. Missions will include fending off attacks on the city's power company, hospital, water system and transportation services.

• The 8th International Conference on emerging Networking EXperiments and Technologies (CoNEXT). Looks like a solid conference, with many interesting papers to follow up on. As seems typical with networking conferences nowadays, the keynote is given by a Ncira employee:

  While many of the original assumptions of SDN have turned out to be less relevant than expected, SDN continues to hold its promise for changing the networks as we know them. I argue the change will not materialize through any individual improvement, new feature, or mechanism but through new abstractions and structures further refining our understanding of network architectures. I will discuss two practically proven examples of such new abstractions and structures – network fabrics and network virtualization – and how they together with software forwarding can help to reorganize functionality within networks, making future networks flexible, evolvable and simple to manage.

• Part II of Tim Roughgarden's Coursera class: Algorithms: Design and Analysis is superb. I think that partly this is because I'm more interested in the material, but partly it's because Professor Roughgarden is clearly much more comfortable with the technology in this second go-round. His lectures are clear and well-paced, his notes are matched, in all ways it just feels like he's comfortable using the tools to teach.

• I love these notes on How to Hack Chipotle

  Basically, you could get two whole avocados at the market for the price of one scoop. Consider ordering a burrito bowl, and adding your sliced avocado on top. Ultimately the guacamole decision is a value call and you have to look deep within your heart and wallet to make the decision that’s right for you. NOTE: If you forgo the meat, the guacamole is free.

  Tell me, though, doesn't it make you think of that classic: The Ultimate In-N-Out Secret Menu (and Super Secret Menu!) Survival Guide?

  Things just got a whole lot more fun. We proceeded to spend the next 15 minutes poring over our options, colluding like '80s kids in a clubhouse trading Garbage Pail Kids, expanding my original list with Thomas' insider information.

• Sometimes it turns out that the island isn't there after all: No Land Ho: Sandy Island and the Age of Un-Discovery

  A phantom island can be defined as ‘An island once believed to exist, and accordingly depicted on maps, but of which the existence was later disproved, and its cartographic representation removed’.

• This looks intriguing: Scribe: The Deterministic Transparent Record/Replay Engine

  Deterministic application record and replay is the ability to record application execution and deterministically replay it at a later time.

• The ever-alert Raymond Chen reminds us that performance regression tests can provide two very different types of information, both of which are valuable: When studying performance, you need to watch out not only for performance degradation, but also unexpected performance improvement

  The second is the one that you don't like: The unexplained improvement. The memory usage activity went down a lot, but you don't remember making any changes that affect your program's memory usage profile. Things got better but you don't know why.

  The danger here is that the performance gain may be the result of a bug. Maybe the scenario completed with half the I/O activity because the storage system is ignoring flush requests. Or it completed 15% faster because the cache is returning false cache hits.