While numerous studies have examined the macro-level behav-ior of traffic in data center networks—overall flow sizes, destina-tion variability, and TCP burstiness—little information is available on the behavior of data center traffic at packet-level timescales. Whereas one might assume that flows from different applications fairly share available link bandwidth, and that packets within a sin-gle flow are uniformly paced, the reality is more complex. To meet increasingly high link rates of 10 Gbps and beyond, batching is typically introduced across the network stack—at the application, middleware, OS, transport, and NIC layers. This batching results in short-term packet bursts, which have implications for the design and performance requireme...
Abstract—Novel packet I/O engines allow capturing traf-fic at multi-10Gb/s using only-software and c...
The multiplexing of variable bit rate traffic streams in a packet network gives rise to two types of...
The goal of this study is to investigate system bottlenecks for high bandwidth applications and how ...
Buffer sizing is a tricky task --- it depends on a large number of variables, ranging from congestio...
We study the burstiness of TCP flows at the packet level. We aggregate packets into entities we call...
Large network operators have thousands or tens of thousands of access aggregation links that they ne...
This memory reference pattern is typical of several types of access in shared-memory systems. Cache-...
Microbursts are traffic events that can cause severe performance degradation in a network. With the ...
Traffic studies is an important subject to evaluate the performance of the switching elements in an ...
Traffic studies is an important subject to evaluate the performance of the switching elements in an ...
this paper, we present an analysis of four traces collected by NLANR, uncovering the characterist...
Abstract-Cloud data centers host diverse applications, mixing in the same network a plethora of work...
Debugging faults in complex networks often requires cap-turing and analyzing traffic at the packet l...
Data centers increasingly deploy commodity servers with high-speed network interfaces to enable low-...
Network interface cards (NICs) are fundamental components of modern high-speed networked systems, su...
Abstract—Novel packet I/O engines allow capturing traf-fic at multi-10Gb/s using only-software and c...
The multiplexing of variable bit rate traffic streams in a packet network gives rise to two types of...
The goal of this study is to investigate system bottlenecks for high bandwidth applications and how ...
Buffer sizing is a tricky task --- it depends on a large number of variables, ranging from congestio...
We study the burstiness of TCP flows at the packet level. We aggregate packets into entities we call...
Large network operators have thousands or tens of thousands of access aggregation links that they ne...
This memory reference pattern is typical of several types of access in shared-memory systems. Cache-...
Microbursts are traffic events that can cause severe performance degradation in a network. With the ...
Traffic studies is an important subject to evaluate the performance of the switching elements in an ...
Traffic studies is an important subject to evaluate the performance of the switching elements in an ...
this paper, we present an analysis of four traces collected by NLANR, uncovering the characterist...
Abstract-Cloud data centers host diverse applications, mixing in the same network a plethora of work...
Debugging faults in complex networks often requires cap-turing and analyzing traffic at the packet l...
Data centers increasingly deploy commodity servers with high-speed network interfaces to enable low-...
Network interface cards (NICs) are fundamental components of modern high-speed networked systems, su...
Abstract—Novel packet I/O engines allow capturing traf-fic at multi-10Gb/s using only-software and c...
The multiplexing of variable bit rate traffic streams in a packet network gives rise to two types of...
The goal of this study is to investigate system bottlenecks for high bandwidth applications and how ...