Protocol Responsibility Offloading to Improve TCP Throughput in Virtualized Environments

Gamage, Sahan; Kompella, Ramana Rao; Xu, Dongyan; Kangarlou, Ardalan

doi:10.1145/2518037.2491463

Cited by 12 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…TCP throughput degradations due to sharing the CPU between cloud VMs are discussed also in [26]. In [27], the authors identify the CPU as one of the bottlenecks that need to be overcome for data transfers.…”

Section: B Copyandcompute Vs Streamandcomputementioning

confidence: 99%

Evaluating Streaming Strategies for Event Processing Across Infrastructure Clouds

Tudoran

Keahey²,

Riteau³

et al. 2014

2014 14th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing

View full text Add to dashboard Cite

Abstract-Infrastructure clouds revolutionized the way in which we approach resource procurement by providing an easy way to lease compute and storage resources on short notice, for a short amount of time, and on a pay-as-you-go basis. This new opportunity, however, introduces new performance trade-offs. Making the right choices in leveraging different types of storage available in the cloud is particularly important for applications that depend on managing large amounts of data within and across clouds. An increasing number of such applications conform to a pattern in which data processing relies on streaming the data to a compute platform where a set of similar operations is repeatedly applied to independent chunks of data. This pattern is evident in virtual observatories such as the Ocean Observatory Initiative, in cases when new data is evaluated against existing features in geospatial computations or when experimental data is processed as a series of time events. In this paper, we propose two strategies for efficiently implementing such streaming in the cloud and evaluate them in the context of an ATLAS application processing experimental data. Our results show that choosing the right cloud configuration can improve overall application performance by as much as three times.

show abstract

Section: B Copyandcompute Vs Streamandcomputementioning

confidence: 99%

Evaluating Streaming Strategies for Event Processing Across Infrastructure Clouds

Tudoran

Keahey²,

Riteau³

et al. 2014

2014 14th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing

View full text Add to dashboard Cite

show abstract

“…Examples of transformed ideas in this space include rethinking TCP performance in virtualized environments [13] and the Route Bricks approach to software router design [8].…”

Section: Programming Languagesmentioning

confidence: 99%

Life in the Fast Lane

Varghese

2015

SIGCOMM Comput. Commun. Rev.

View full text Add to dashboard Cite

The most striking ideas in systems are abstractions such as virtual memory, sockets, or packet scheduling. Algorithmics is the servant of abstraction, allowing the performance of the system to approach that of the underlying hardware. I survey the trajectory of network algorithmics, starting with a focus on speed and scale in the 1990s to measurement and security in the 2000s, using what I call the confluence lens.Confluence sees interdisciplinary work as a merger of two or more disciplines made compelling by an inflection point in the real world, while also producing genuinely transformed ideas. I attempt to show that Network Algorithmics represented a confluence in the 1990s between computer systems, algorithms, and networking. I suggest Confluence Diagrams as a means to identify future interdisciplinary opportunities, and describe the emerging field of Network Verification as a new confluence between networking and programming languages.

show abstract

“…Multiple researchers expose differences in latency and throughput in virtualized networking devices [19], [23], [25], [27]. Virtualization can also cause different network behaviors, especially around TCP and congestion control [9], [12], [13]. These papers typically present new approaches for performance improvement.…”

Section: Introductionmentioning

confidence: 99%

Virtually the Same: Comparing Physical and Virtual Testbeds

Crussell

Kroeger

Brown

et al. 2019

2019 International Conference on Computing, Networking and Communications (ICNC)

View full text Add to dashboard Cite

Network designers, planners, and security professionals increasingly rely on large-scale testbeds based on virtualization to emulate networks and make decisions about real-world deployments. However, there has been limited research on how well these virtual testbeds match their physical counterparts. Specifically, does the virtualization that these testbeds depend on actually capture real-world behaviors sufficiently well to support decisions?As a first step, we perform simple experiments on both physical and virtual testbeds to begin to understand where and how the testbeds differ. We set up a web service on one host and run ApacheBench against this service from a different host, instrumenting each system during these tests.We define an initial repeatable methodology (algorithm) to quantitatively compare physical and virtual testbeds. Specifically we compare the testbeds at three levels of abstraction: application, operating system (OS) and network. For the application level, we use the ApacheBench results. For OS behavior, we compare patterns of system call orderings using Markov chains. This provides a unique visual representation of the workload and OS behavior in our testbeds. We also drill down into read-system-call behaviors and show how at one level both systems are deterministic and identical, but as we move up in abstractions that consistency declines. Finally, we use packet captures to compare network behaviors and performance. We reconstruct flows and compare per-flow and per-experiment statistics.From these comparisons, we find that the behavior of the workload in the testbeds is similar but that the underlying processes to support it do vary. The low-level network behavior can vary quite widely in packetization depending on the virtual network driver. While these differences can be important, and knowing about them will help experiment designers, the core application and OS behaviors still represent similar processes.

show abstract

Protocol Responsibility Offloading to Improve TCP Throughput in Virtualized Environments

Cited by 12 publications

References 0 publications

Evaluating Streaming Strategies for Event Processing Across Infrastructure Clouds

Evaluating Streaming Strategies for Event Processing Across Infrastructure Clouds

Life in the Fast Lane

Virtually the Same: Comparing Physical and Virtual Testbeds

Contact Info

Product

Resources

About