"Dynamo" is a wonderful introduction to distributed systems in general, and is a remarkably readable paper overall. It was the first paper I read as a "professional", and I think holds that status for many. https://s3.amazonaws.com/systemsandpapers/papers/amazon-dynamo-sosp2007.pdf …
-
Show this thread
-
"Hints for Computer System Design" is amazing, in part because it was written in '83! The paper is a survey of learnings from Butler Lampson's design of many systems, on interface design, premature optimization, and many more. https://s3.amazonaws.com/systemsandpapers/papers/acrobat-17.pdf …
1 reply 0 retweets 3 likesShow this thread -
It also features this lethal quote from C Hoare, "Algol 60 was not only an improvement on its predecessors, but also on nearly all its successors." Shade thrown.
1 reply 0 retweets 3 likesShow this thread -
"Big Ball of Mud" is a reaction against papers celebrating grandiose design patterns, arguing that the defining pattern for most applications is THE BIG BALL OF MUD. Why is this pattern so popular, if we all agree it's a tire fire? https://s3.amazonaws.com/systemsandpapers/papers/bigballofmud.pdf …
1 reply 0 retweets 3 likesShow this thread -
"The Google File System" looks at GFS, which was potentially *the* defining infrastructure abstraction that underpinned Google's systems until the development of Colossus. (Colossus is really cool to read about too!) https://s3.amazonaws.com/systemsandpapers/papers/gfs.pdf …
2 replies 0 retweets 5 likesShow this thread -
"On Designing and Deploying Internet-Scale Services" is James Hamilton dropping a huge amount of wisdom around how to run large-scale systems, even more impressive for being written in '07 and I still learn more each time I read it. https://s3.amazonaws.com/systemsandpapers/papers/hamilton.pdf …
1 reply 1 retweet 10 likesShow this thread -
I mean, how many infrastructure teams in 2018 are tracking their "system-to-administrator ratio" as Hamilton discusses? Do you have a target ratio that you're trying to hit? (Me? Oh, yeah, of course I do... eek.)
1 reply 0 retweets 3 likesShow this thread -
"Harvest, Yield and Scalable Tolerant Systems" is a nuanced look at CAP, and how the vocabulary how breaks down in discussing how systems degrade. h/t to
@kiranb who introduced me to this one! https://s3.amazonaws.com/systemsandpapers/papers/FOX_Brewer_99-Harvest_Yield_and_Scalable_Tolerant_Systems.pdf …2 replies 0 retweets 7 likesShow this thread -
Ok, and there are like twenty other papers as well! What are some of your favorite papers?
14 replies 0 retweets 1 likeShow this thread -
Replying to @Lethain
A couple of nice surveys of ML sub-topics: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.297.3086 … https://arxiv.org/pdf/1710.06068.pdf … A neat security application of probabilistic data structures: https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/popularityISeverything.pdf … ML eng best practices: http://martin.zinkevich.org/rules_of_ml/rules_of_ml.pdf …
2 replies 1 retweet 6 likes
Also word2vec is pretty amazing: https://arxiv.org/pdf/1301.3781.pdf … And later work on word embeddings gets even crazier: https://arxiv.org/abs/1710.04087
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.