Zobrazeno 1 - 10
of 28
pro vyhledávání: '"Dushyanth Narayanan"'
Autor:
Jiaqi Chu, Nathanaël Cheriere, Grace Brennan, Mengyang Yang, Greg O’Shea, Jannes Gladrow, Douglas J. Kelly, Giorgio Maltese, Alan Sanders, Dushyanth Narayanan, Benn Thomsen, Antony Rowstron
Publikováno v:
Communications Engineering, Vol 3, Iss 1, Pp 1-9 (2024)
Abstract Cloud data workloads require both high capacity at low cost and high access rates. Hard Disk Drives are the dominant media in this application as they are low cost, however, Hard Disk Drive technology is seeing declining access rates and a s
Externí odkaz:
https://doaj.org/article/d3e0c0c39a8a4524969f65ebd601a69a
Publikováno v:
FAST
In enterprise data centers power usage is a problem impacting server density and the total cost of ownership. Storage uses a significant fraction of the power budget and there are no widely deployed power-saving solutions for enterprise storage syste
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::db93c1bdcbfaa7fc0df276ad826cc6f5
https://doi.org/10.1145/1416944.1416949
https://doi.org/10.1145/1416944.1416949
Publikováno v:
The VLDB Journal. 17:315-331
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::f91193ea6f029401c0ef3e721a928808
https://doi.org/10.1007/s00778-007-0060-3
https://doi.org/10.1007/s00778-007-0060-3
Autor:
Dushyanth Narayanan, Matthew Renzelmann, Aleksandar Dragojevic, Anirudh Badam, Alex Shamis, Miguel Castro, Edmund B. Nightingale
Publikováno v:
SOSP
Transactions with strong consistency and high availability simplify building and reasoning about distributed systems. However, previous implementations performed poorly. This forced system designers to avoid transactions completely, to weaken consist
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::42abd359779b9bf66de5e0f9f4d4cce1
https://doi.org/10.1145/2815400.2815425
https://doi.org/10.1145/2815400.2815425
Publikováno v:
Wireless Networks. 7:601-607
We introduce the concept of multi-fidelity algorithms, which revises the classical notion of an algorithm. Instead of having a fixed output criterion and allowing the resource consumption to vary, we bound the resource consumption and allow the fidel
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::e4e78c3cf1c8d08fa30666ec7cd18dd8
https://doi.org/10.1023/a:1012315301312
https://doi.org/10.1023/a:1012315301312
Publikováno v:
SoCC
In the last decade we have seen a huge deployment of cheap clusters to run data analytics workloads. The conventional wisdom in industry and academia is that scaling out using a cluster of commodity machines is better for these workloads than scaling
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::fc2fa11d2044d87e71f8f8bb8c4df030
https://doi.org/10.1145/2523616.2523629
https://doi.org/10.1145/2523616.2523629
Publikováno v:
[Research Report] RR-8188, INRIA. 2012, pp.24
HAL
HAL
As a result of continuous innovation in hardware technology, computers are made more and more powerful than their prior models. Modern servers nowadays can possess large main memory capability that can size up to 1 Terabytes (TB) and more. As memory
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::e70d32e4cb3846d42f7e11df2a9efd09
https://hal.inria.fr/hal-00766219
https://hal.inria.fr/hal-00766219
Publikováno v:
Proceedings of the 1st International Workshop on Hot Topics in Cloud Data Processing.
The norm for data analytics is now to run them on commodity clusters with MapReduce-like abstractions. One only needs to read the popular blogs to see the evidence of this. We believe that we could now say that "nobody ever got fired for using Hadoop
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::84839dbce7956897ca99cb8efee92794
https://doi.org/10.1145/2169090.2169092
https://doi.org/10.1145/2169090.2169092
Autor:
Dushyanth Narayanan, Orion Hodson
Publikováno v:
ASPLOS
Today's databases and key-value stores commonly keep all their data in main memory. A single server can have over 100 GB of memory, and a cluster of such servers can have 10s to 100s of TB. However, a storage back end is still required for recovery f
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d5725e3f095ba95cd4a60159f1178359
https://doi.org/10.1145/2150976.2151018
https://doi.org/10.1145/2150976.2151018
Publikováno v:
EuroSys
Online services hosted in data centers show significant diurnal variation in load levels. Thus, there is significant potential for saving power by powering down excess servers during the troughs. However, while techniques like VM migration can consol
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::54e8890185dfa0faff70b00e2d1fe6c2
https://doi.org/10.1145/1966445.1966461
https://doi.org/10.1145/1966445.1966461