∫ntegrabℓε ∂iﬀerentiαℓs · unorganised notes, code, and writings of random topics

Vasudevan et al (2009) A (In)Cast of Thousands: Scaling Datacenter TCP to Kiloservers and Gigabits (CMU TR)

July 8, 2011 paper

A tech report to study the incast behaviour in data center. Incast is due to RTO in synchronized read. To solve the problem, set the RTOmin to 200us can help. The default value of 200ms gives a poor goodput. [more]

Papers related to TCP research

July 6, 2011 blog work

Some of the TCP papers I read in recent years, with short notes. [more]

The case of back door in vsftpd 2.3.4

July 3, 2011 blog code

Source: http://pastebin.com/AetT9sS5, http://scarybeastsecurity.blogspot.com/2011/07/alert-vsftpd-download-backdoored.html [more]

Ren et al (2010) Google-Wide Profiling: A Continuous Profiling Infrastructure for Data Centers (IEEE Micro)

June 21, 2011 paper

Google-Wide Profiling is theoretically an extension of the Digital Continuous Profiling Infrastructure (DCPI) [Anderson et al, 1997] to data centers. It is a continuous profiling infrastructure that samples across machines in multiple data centers and collects various events. GWP collects profiles on stack traces, hardware events, lock contention profiles, heap... [more]

Integration tricks

June 3, 2011 blog math

Source [more]