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

Log of numbers

September 8, 2010 blog math

It is well known to some of us: When you are dealing with numbers in computer, the floating point arithmetics makes you lost the precision. For example, adding a bunch of small numbers (e.g. a million of millionth) to a large number never gives an accurate result, unless you use... [more]

Kabbani (2010) AF-QCN: Approximate Fairness with Quantized Congestion Notification for Multi-tenanted Data Centers (HotI)

September 6, 2010 paper

Consider a data center network using CEE (802.1Qau) with different tenants. If a tenant is running way larger number of flows, according to 802.1Qau, he can get most of the bandwidth because 802.1Qau only enforces per-flow fairness (even distribution) of bandwidth allocation. We cannot do (1) per-user even distribution and... [more]

Isard et al (2009) Quincy: Fair Scheduling for Distributed Computing Clusters (SOSP)

September 6, 2010 paper

A paper from Microsoft Research Silicon Valley. [more]

Al-Fares et al (2010) Hedera: Dynamic Flow Scheduling for Data Center Networks (NSDI)

August 2, 2010 paper

This is about flow scheduling for multiroot networks using centralized server. [more]

Wang et al (2010) c-Through: Part-time Optics in Data Centers (SIGCOMM)

August 2, 2010 paper

This paper proposed a term, HyPaC network, for hybrid of packet and circuit switched network. The packet switched network are commodity electrical ethernet switch and the circuit switched network is optical connections. This paper built a HyPaC testbed, c-through. [more]