A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Filters
Towards causal datacenter networks
2018
Proceedings of the 5th Workshop on the Principles and Practice of Consistency for Distributed Data - PaPoC '18
Preserved Fulltext
This approach, which we call omnisequencing, would ensure causal delivery of all messages, making consistency a network-level guarantee. ...
We build on that trend and ask whether it is possible for the datacenter network to order all messages in a protocol-agnostic way. ...
A causally ordered datacenter network resembles traditional software-based primitives like causal multicast from ISIS [1] , but can be implemented efficiently and at large scale. ...
doi:10.1145/3194261.3194269
dblp:conf/eurosys/MichaelP18
fatcat:aundnveep5fwhoasmsn3vseyl4
YTrace: End-to-end Performance Diagnosis in Large Cloud and Content Providers
[article]
2016
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
We show three case studies from production that span a large-scale distributed storage system, a datacenter-wide network, and an end-to-end video serving stack at Yahoo. ...
YTrace diagnoses problems across service and network layers in the end-to-end path spanning user host, Internet, CDN and the datacenters, and has three diagnosis goals: detection, localization and root ...
We first look at datacenter network problem diagnosis. Datacenter network diagnosis. ...
arXiv:1602.03273v3
fatcat:7ufshgqpfzh4xiott5ve53wwoa
GentleRain
2014
Proceedings of the ACM Symposium on Cloud Computing - SOCC '14
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
By encoding causality in a single scalar timestamp, GentleRain reduces storage and communication overhead for tracking causality. ...
GentleRain is a new causally consistent geo-replicated data store that provides throughput comparable to eventual consistency and superior to current implementations of causal consistency. ...
This leaves network failures both within and across datacenters. ...
doi:10.1145/2670979.2670983
dblp:conf/cloud/DuI0Z14
fatcat:afphmev7ibal5ajxkpwfbhixse
The Challenges of Global-scale Data Management
2016
Proceedings of the 2016 International Conference on Management of Data - SIGMOD '16
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
However, the Wide-Area Network (WAN) latency separating data is orders of magnitude larger than conventional network latencies, and this requires a reevaluation of many of the traditional design trade-offs ...
Leveraging advances in WAN research from the networking community is an important step towards building efficient GSDM systems. ...
Round-Trip Time
communication (WAN RTT) compared to memory access la-
tency [50] and network latency within the datacenter (local RTT) ...
doi:10.1145/2882903.2912571
dblp:conf/sigmod/NawabAA16
fatcat:4ja2phoh7zajxnckcupa25o65y
Challenges facing tomorrow's datacenter
2008
Proceedings of the 2nd Workshop on Large-Scale Distributed Systems and Middleware - LADIS '08
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2012; you can also visit the original URL.
The file type is application/pdf.
The 2008 workshop on Large-Scale Distributed Systems and Middleware (LADiS) addressed challenges facing tomorrow's datacenter. ...
Doing so is useful for structuring an overlay mesh, with high capacity nodes forming the backbone of the mesh, and low capacity nodes moved toward the edges of the network. ...
The failure and network characteristics within a datacenter often drastically differ from those across datacenters. ...
doi:10.1145/1529974.1529976
fatcat:zmlphu6l6ndovnxur5znfugh24
Distributed Cloud Computing
2015
Computer communication review
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2015; you can also visit the original URL.
The file type is application/pdf.
A distributed cloud connecting multiple, geographically distributed and smaller datacenters, can be an attractive alternative to today's massive, centralized datacenters. ...
His talk then presented a survey on some practical first steps towards addressing these challenges. ...
Distributed cloud computing enables the moving of execution of distributed applications towards client machines. ...
doi:10.1145/2766330.2766337
fatcat:girpjoy7cbfyzbgylqa77d47cy
2019 Index IEEE Transactions on Services Computing Vol. 12
2020
IEEE Transactions on Services Computing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
-Feb. 2019 105-116 Minimizing Traffic Migration During Network Update in IaaS Datacenters. ...
-Feb. 2019 105-116 Minimizing Traffic Migration During Network Update in IaaS Datacenters. ...
doi:10.1109/tsc.2020.2965435
fatcat:tim4rhxag5dqpbhvtgpwvp6ibm
An Adaptive Replica Selection Algorithm for Quorum based Distributed Storage System
2016
International Journal of Grid and Distributed Computing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
In Quorum based cross datacenter distributed storage systems, data replication is a widely used technology. ...
Network latency between different datacenters has huge difference, thus in cross datacenter scenario this performance reduction issue is more obvious. ...
We use 7 desktops to simulate 7 datacenters. Each machine contains a Cassandra node. The network latencies between different datacenters are simulated by adding some extra delay time. ...
doi:10.14257/ijgdc.2016.9.5.06
fatcat:nnbbbwpcnjffpapejrh2b7wmny
Delay-Based Cloud Congestion Control
2009
GLOBECOM 2009 - 2009 IEEE Global Telecommunications Conference
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
A qualitative open loop analysis of two datacenter networks (10 Gb/s Ethernet and 12x Infiniband) yields positive results. ...
To address this problem we first investigate whether path delay could serve as a reliable congestion measure for clouds spanning across multiple datacenters and heterogenous networks. ...
CLOUD AND DATACENTER TRENDS Cost and power constrains are leading toward IT federation, i.e., the aggregation of server blades, clusters, and storage networks into increasingly larger datacenters (DC). ...
doi:10.1109/glocom.2009.5425643
dblp:conf/globecom/GusatBM09
fatcat:zk6sdcai4jbpjboohrkf2je44y
Distributed Data-Centric Adaptive Sampling for Cyber-Physical Systems
2015
ACM Transactions on Autonomous and Adaptive Systems
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2016; you can also visit the original URL.
The file type is application/pdf.
The causality analysis and the same correlations are then leveraged for adaptive sleep scheduling aimed at saving energy in wireless sensor networks (WSNs). ...
Differently from existing adaptive sampling solutions, SILENCE employs temporal causality analysis to not only track the variation in the underlying phenomenon but also its cause and direction of propagation ...
Part (a) shows the causality flow (sign and magnitude), whereas parts (b) and (c) show the causality information in network form and matrix form, respectively. ...
doi:10.1145/2644820
fatcat:wqao3gyl7rhjjhtmra4krmbuam
Got loss?
2013
Proceedings of the ACM SIGCOMM 2013 conference on SIGCOMM - SIGCOMM '13
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Datacenter networking is currently dominated by two major trends. ...
Moreover, we propose a zero-loss Overlay Virtual Network (zOVN) designed to reduce the query and flow completion time of latency-sensitive datacenter applications. ...
The "noise" in Figure 3c confirms our intuitive hypothesis about large non-causal performance variability in virtual networks. ...
doi:10.1145/2486001.2486024
dblp:conf/sigcomm/CrisanBCMG13
fatcat:c4culop7lbfr3hrqwvpumw3tnm
Improved routing algorithms in the dual-port datacenter networks HCN and BCN
2017
Future generations computer systems
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
We also present a much tighter analysis of HCN and BCN by observing that there is a very close relationship between the datacenter networks HCN and the interconnection networks known as WK-recursive networks ...
Highlights Improved routing algorithms for the datacenter networks HCN and BCN are proposed. New routing algorithms are derived from algorithms for WK-recursive networks. ...
These new workloads cover more representative and realistic network traffic scenarios that can be found in existing datacenters. ...
doi:10.1016/j.future.2017.05.004
fatcat:ojcarxzokvdcbpelwgqspiqgvu
Revisiting Time, Clocks, and Synchronization
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Finally, we briefly discuss an independent time network approach towards low-cost and high-precision synchronization. ...
Sub-nanosecond precision clock synchronization over the packet network has been achieved by the White Rabbit protocol for a decade. However, few computer systems utilize such a technique. ...
We proposed an approach of deploying an independent time network. This paper is only a vague start. We hope to attract more interest in the clock synchronization problem. ...
arXiv:2106.16140v1
fatcat:o6wf7zx6ozb5zm74zbl77pd2uq
Automating the debugging of datacenter applications with ADDA
2013
2013 43rd Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
Debugging data-intensive distributed applications running in a datacenter ("datacenter applications") is complex and time-consuming. ...
We present ADDA, a replay-debugging system for datacenter applications. ...
Unfortunately, debugging datacenter applications is hard. When an application failure occurs, the causality chain of the failure is often difficult to trace, as it may span multiple nodes. ...
doi:10.1109/dsn.2013.6575303
dblp:conf/dsn/ZamfirAS13
fatcat:cmb7j7vpkvhfxdvr5dvqxjsb5m
The Datacenter as a Computer: An Introduction to the Design of Warehouse-Scale Machines, Second edition
2013
Synthesis Lectures on Computer Architecture
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Their approach consists of observing networking traffic among system components and inferring causal relationships through statistical inference methods. 29 Because they treat all system components ( ...
Typically, approximately 80% of total construction cost goes toward power and cooling, and the remaining 20% toward the general building and site construction. ...
doi:10.2200/s00516ed2v01y201306cac024
fatcat:435o455inbcmrakl6l7jp4gope
« Previous
Showing results 1 — 15 out of 435 results