Performance Analysis of Large Receive Offload in a Xen Virtualized System

Oi, Hitoshi; Nakajima, Fumio

doi:10.1109/iccet.2009.112

Cited by 14 publications

(10 citation statements)

References 5 publications

(6 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is because we only run the guest domain for the message receiver on the receiver machine to see the effects of LRO and optimizations in this paper. Changes in the network performance when another guest domain is consuming CPU time can be found in [10] 1 . The base operating system is Linux kernel 2.6.18 and Xen 3.1.1 is used for virtualization.…”

Section: Experimental Results and Analysismentioning

confidence: 99%

“…To obtain the packet handling overhead in number of clock cycles, we use oprofile [12]. For the measurements of native Linux, we actually use the same system as Xen cases, but the netserver (the receiver program of the netperf suite) is run inside the driver domain so that the results are not affected by the 1 Optimizations discussed in this paper were not applied in [10] overhead of Xen internal network architecture described in Section II-A. …”

Section: Experimental Results and Analysismentioning

confidence: 99%

“…Each test is run for five minutes. For the TCP STREAM TEST, a single message size of 1KB is used because, from our experiences in [10], the throughput is almost constant for the message sizes of 256B or larger. To obtain the packet handling overhead in number of clock cycles, we use oprofile [12].…”

Section: Experimental Results and Analysismentioning

confidence: 99%

“…In our previous work, we ported the skb-mode LRO into the physical and virtual NICs of a Xen virtualized system and measured its effectiveness [9], [10]. In the previous work, we noticed of two possible points of further optimization.…”

Section: Further Optimizationsmentioning

confidence: 99%

See 3 more Smart Citations

Optimizations of Large Receive Offload in Xen

Nakanjima

2009

2009 Eighth IEEE International Symposium on Network Computing and Applications

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Experimental Results and Analysismentioning

confidence: 99%

Section: Experimental Results and Analysismentioning

confidence: 99%

Section: Experimental Results and Analysismentioning

confidence: 99%

Section: Further Optimizationsmentioning

confidence: 99%

See 2 more Smart Citations

Optimizations of Large Receive Offload in Xen

Nakanjima

2009

2009 Eighth IEEE International Symposium on Network Computing and Applications

Self Cite

View full text Add to dashboard Cite

show abstract

“…15,16,17,18,19,20,21,22, and 23, we present the measured metrics details on Platform II, to which the previous similar analysis approaches on Platform I could also be applied. Most previous conclusions still work under Platform II, only except the instance of two CPU-intensive workloads combination (i.e., (1K, 1K)).…”

Section: Platform Ii: Throughput Interference Analysismentioning

confidence: 99%

Who Is Your Neighbor: Net I/O Performance Interference in Virtualized Clouds

Liu

Mei

et al. 2013

IEEE Trans. Serv. Comput.

View full text Add to dashboard Cite

Abstract-User-perceived performance continues to be the most important QoS indicator in cloud-based data centers today. Effective allocation of virtual machines (VMs) to handle both CPU intensive and I/O intensive workloads is a crucial performance management capability in virtualized clouds. Although a fair amount of researches have dedicated to measuring and scheduling jobs among VMs, there still lacks of in-depth understanding of performance factors that impact the efficiency and effectiveness of resource multiplexing and scheduling among VMs. In this paper, we present the experimental research on performance interference in parallel processing of CPU-intensive and network-intensive workloads on Xen virtual machine monitor (VMM). Based on our study, we conclude with five key findings which are critical for effective performance management and tuning in virtualized clouds. First, colocating network-intensive workloads in isolated VMs incurs high overheads of switches and events in Dom0 and VMM. Second, colocating CPU-intensive workloads in isolated VMs incurs high CPU contention due to fast I/O processing in I/O channel. Third, running CPU-intensive and network-intensive workloads in conjunction incurs the least resource contention, delivering higher aggregate performance. Fourth, performance of network-intensive workload is insensitive to CPU assignment among VMs, whereas adaptive CPU assignment among VMs is critical to CPU-intensive workload. The more CPUs pinned on Dom0 the worse performance is achieved by CPU-intensive workload. Last, due to fast I/O processing in I/O channel, limitation on grant table is a potential bottleneck in Xen. We argue that identifying the factors that impact the total demand of exchanged memory pages is important to the in-depth understanding of interference costs in Dom0 and VMM.

show abstract